University of Turku Modern Industrial Materials Master’s Thesis Guide: MTEK0011, 30 ECTS, Industrial Materials and UTUGradu

What this guide covers

This guide is for the University of Turku Modern Industrial Materials track in the current 2024-2027 Peppi curriculum. It uses the exact MTEKMSCMODIND2427, programme 100064 object rather than a generic materials-science thesis model. The key structural point is that the 40 ECTS Thesis and Project category is not a 40 ECTS thesis. It contains MTEK0011 Master’s Thesis in Technology, Materials Engineering, 30 ECTS, TTDK1308 Degree Qualifying Examination, 0 ECTS, and MTEK0019 Capstone, 10 ECTS. MTEK0020 is a separate 5 ECTS Pass/Fail thesis seminar in Common Advanced Studies. The guide also explains how simulation, imaging, manufacturing, metrology, optimisation, machine-learning and lifecycle evidence should be bounded.

Current programme object

The current Peppi object is MTEKMSCMODIND2427, programme 100064. The degree is Master of Science (Technology), organised by the Faculty of Technology and the Department of Mechanical and Materials Engineering. The formal degree size is 120 ECTS. Modern Industrial Materials is one of three Materials Engineering specialisation tracks, so students should use the curriculum period attached to their own study right and HOPS rather than copying an older curriculum or a neighbouring track. Programme codes and category placement matter because the thesis, seminar and Capstone are related in the degree but remain different assessed objects.

How the 120 ECTS degree is structured

The current programme combines 20 ECTS Common Advanced Studies in Materials Engineering, 20 ECTS Modern Industrial Materials studies, a 40 ECTS Thesis and Project category, 20-25 ECTS minor or thematic studies and 15-20 ECTS other studies. Peppi models the advanced-studies side as an 80 ECTS module. The track-specific 20 ECTS block contains MTEK0024 Simulations and New Materials, MTEK0025 Sustainability and Life-Cycle Management of Materials, and 10 ECTS chosen from approved industrial-materials options. This structure is useful when planning workload because the thesis is only one part of the advanced-studies package.

Read the 40 ECTS Thesis and Project block correctly

The 40 ECTS category contains three different academic objects. MTEK0011 is the individually examined 30 ECTS master’s thesis. MTEK0019 is a separate 10 ECTS Capstone course. TTDK1308 is a 0 ECTS maturity examination. The category label therefore must not be turned into a statement that the thesis itself is 40 ECTS. If a Capstone problem, prototype or company collaboration develops into a thesis, the MTEK0011 work still needs its own research question, method, evidence chain, manuscript and examination. Keeping the objects separate also prevents double counting when explaining the project to a supervisor or company.

Exact thesis course: MTEK0011

MTEK0011 Master’s Thesis in Technology, Materials Engineering is 30 ECTS, belongs to Advanced Studies and is graded on the 0-5 scale. The course trains practical and theoretical analysis of research problems through scientific literature and research methods. The thesis should demonstrate knowledge of the field, management of research methods and scientific writing. In Modern Industrial Materials, a broad theme such as additive manufacturing, corrosion, coatings, simulations or machine learning is therefore only a starting area. A workable thesis turns the area into a focused question that can be answered with evidence within the 30 ECTS scope.

The separate MTEK0020 thesis seminar

MTEK0020 Master’s Thesis in Technology Seminar, Materials Engineering is a separate 5 ECTS Advanced Studies course graded Pass/Fail. Peppi places it in Common Advanced Studies rather than inside MTEK0011 or the Thesis and Project category. Students prepare and present a short research plan, analyse another completed thesis, present their own results near completion, participate in discussion and keep the required learning work. The seminar supports the thesis process but does not change MTEK0011 from 30 ECTS to 35 ECTS. Plan the research-plan presentation early enough that seminar feedback can still influence the actual project.

The separate MTEK0019 Capstone

MTEK0019 Capstone is a separate 10 ECTS Advanced Studies course graded 0-5. It addresses real-life challenges through team work, often with companies, communities or research groups. A Capstone can create a prototype, dataset, manufacturing concept or stakeholder problem that later motivates an individual thesis, but the team output is not automatically the thesis evidence. If the same industrial context is used in both courses, agree the boundary with supervisors at the beginning. The MTEK0011 thesis still needs an individually defensible question, method, analysis and examined manuscript.

TTDK1308 maturity examination

TTDK1308 is the 0 ECTS Degree Qualifying Examination for the master’s degree. In the normal route, the thesis abstract or another suitable thesis part can function as the maturity test and demonstrate familiarity with the thesis field. A conditional written-exam route can apply depending on the student’s previous degree and Finnish or Swedish educational-language background. Because the route is student-specific, do not copy another student’s maturity arrangement. Check your own HOPS and the current University instructions before final submission so a completed thesis is not delayed by an unresolved degree-completion requirement.

Finding a topic and supervisor

MTEK0011 directs students to contact a professor of the main subject when seeking a thesis topic so supervision can be assigned. Topics may come from a research group, a company or another organisation. In this track, useful questions can concern material structure, processing, surface degradation, dimensional accuracy, manufacturing routes, model predictions, material selection or lifecycle performance. Avoid beginning with a claim such as a material is simply “better”. Define what property is being improved, compared with what reference, under which conditions, and at which evidence level. A narrow question is easier to supervise, analyse and examine.

Build a thesis plan before experiments or modelling

At the beginning, define the problem, research question, materials or datasets, evidence level, methods, controls, analysis rules, equipment or software, permissions, risks, milestones and expected outputs. Industrial-materials projects can depend on sample availability, manufacturing slots, heat treatment, coating runs, instrument access, simulation time, company data or external laboratories. A written plan makes these dependencies visible before they become delays. It also reduces the temptation to rewrite the research question after seeing a favourable result. If the plan changes, record why the change was methodologically necessary and what it means for the original comparison.

Use supervision as evidence control

Regular supervision is not only a progress meeting. Use it to resolve decisions that affect the evidence: failed samples, altered processing parameters, calibration problems, excluded measurements, image-selection rules, simulation assumptions, optimisation constraints, unusual outliers, data-cleaning rules and changes in scope. Keep a decision record so the Methods section can describe what actually happened rather than an idealised procedure. If a project moves from material characterisation to component performance, from computational prediction to manufacturing claims, or from technical results to sustainability conclusions, confirm that the new claim level is supported before widening the thesis conclusion.

How examination and grading work

The current MTEK0011 description states that at least two examiners evaluate the thesis under University evaluation and grading guidance. Final acceptance is decided by the head of the department, with the grade based on examiner evaluation. Supervisor support is therefore not the final examination decision. Write so an examiner can trace the research question through literature, method, results, uncertainty and conclusion. A striking micrograph, high optimisation score, accurate ML model or successful printed component does not compensate for an unclear comparison, undocumented method or a conclusion that goes beyond the evidence actually generated.

Choose a research question at the correct evidence level

Modern Industrial Materials work can move across atomic structure, microstructure, surface state, material property, manufactured geometry, component performance, process capability, economic feasibility and lifecycle impact. These levels are related but not interchangeable. A simulated elastic property does not automatically prove the same property in a manufactured part. A better surface finish does not itself prove longer service life. A high ML score does not prove a physical mechanism. State the intended evidence level in the research question and keep the main conclusion at that level unless the project deliberately adds the methods needed to support a broader claim.

Organise the literature around the evidence chain

A strong literature review should show how prior work connects processing, structure, property and application. Separate experimental characterisation from simulations, manufacturing studies, metrology, optimisation, predictive modelling and lifecycle analysis. When comparing published properties, check material composition, heat treatment, specimen geometry, loading mode, environment, measurement method and uncertainty. Avoid ranking materials from headline numbers produced under incompatible conditions. The review should explain what is directly comparable, what needs normalisation or caution, which mechanisms are established, and which proposed relationships remain hypotheses that your thesis can test.

Simulations and New Materials: define what is computational

MTEK0024 Simulations and New Materials connects material structure and bonding with atomistic simulation, material properties and possible device applications. A computational thesis should state the simulated structure, code, method, boundary conditions, parameter choices and target property. Distinguish a calculated result from a measured result throughout the manuscript. If the project predicts that a modified structure may improve a functional property, present this as a computational conclusion unless independent experiments validate it. The useful contribution can still be substantial: a well-bounded simulation can explain trends, screen alternatives or identify hypotheses for later experimental work without pretending to be physical qualification.

Multiscale modelling needs explicit assumptions

MTEK0033 Multiscale Modelling emphasises that different material properties operate at different spatial and temporal scales and may require different modelling approaches. Record the assumptions, simplifications and scale of the chosen model. A finite-element model, particle-based simulation and atomistic model answer different questions. Explain why the selected approach is appropriate for the property of interest and which phenomena are omitted. If parameter values come from literature or separate experiments, identify their source. Sensitivity analysis is especially useful when uncertain input values materially affect the result, because it shows whether the conclusion is robust or only valid under one parameter choice.

Validate models at the level of the intended claim

Validation should match the claim. If the thesis claims that a model reproduces a measured deformation response, compare the model with suitable experimental data and describe the comparison metric. If only qualitative agreement is available, say so. Do not validate one output and imply that every model output is therefore correct. Calibration and validation should be distinguished when the same dataset is used to tune parameters. Extrapolation outside the studied loading, temperature, composition or geometry range should be labelled as prediction. This makes a computational thesis stronger because readers can see where evidence ends and where model-dependent expectation begins.

Imaging methods should answer the research question

MTEK0034 covers optical microscopy, SEM, EDS/EDX, AFM, TEM and other imaging approaches. Select an imaging method because it addresses a property or structure relevant to the research question, not simply because the instrument is available. Morphology, topography, elemental signal and fine structural information are different evidence. Report sample preparation, imaging conditions and selection rules. If the thesis makes a quantitative claim, use a reproducible sampling and analysis procedure rather than relying on one attractive image. When comparing processed and reference samples, keep preparation and measurement conditions sufficiently consistent to avoid creating an artificial difference.

Do not confuse morphology, composition and mechanism

An SEM image can show morphology but does not by itself establish chemical composition. An EDS signal can support elemental interpretation but does not automatically prove a particular phase or reaction mechanism. AFM topography is not a substitute for chemistry, and a TEM image may represent a highly selected local region. When mechanism matters, combine evidence where possible and explain how each method contributes. If the project demonstrates only an association between processing and morphology, report that association honestly. Causal language should be reserved for designs and converging evidence that can support it.

Surface and coating studies need a defined degradation problem

KTEK0035 covers coatings, mechanical and thermal surface engineering, wear and corrosion. A thesis in this area should define the substrate, coating or treatment, environment, loading and degradation mechanism relevant to the application. A coating that performs well under one abrasion or corrosion test may not behave the same way under a different temperature, electrolyte, stress state or combined mechanism. Report preparation and thickness where relevant, identify the reference condition, and distinguish surface performance from bulk-material performance. This keeps conclusions tied to the actual test rather than converting a laboratory improvement into a universal durability claim.

Service-life claims require more than one favourable test

Accelerated corrosion, wear or thermal tests can be valuable for comparing material states, but accelerated conditions may not reproduce every mechanism present in service. If the thesis estimates lifetime, explain the relationship between the test and the intended operating condition. Avoid converting a short accelerated result directly into years of service without a validated model or justified basis. Where long-term validation is outside a 30 ECTS project, frame the result as comparative degradation behaviour or early evidence. A bounded conclusion is more defensible than an unsupported lifetime number and can still be highly useful for material selection or follow-up research.

Digital manufacturing: connect process, microstructure and property

KTEK0034 Materials Engineering in Digital Manufacturing covers material classes, solidification, microstructure, powder metallurgy, sintering, feedstock characteristics, heat treatment and hot isostatic pressing. These variables can interact, so document the processing history of each compared sample. If a thesis claims that a process improves a mechanical or functional property, establish the property with an appropriate measurement rather than inferring it from microstructure alone. Conversely, if the purpose is to explain a microstructure change, do not overstate the result as full component performance. A clear process-structure-property chain helps readers understand which link the thesis actually tested.

Manufacturing-process advantages and limitations are application-specific

KTEK0033 covers modern manufacturing routes, including welding, laser-based processes and thermal spray, together with process advantages and limitations. Avoid statements that one process is universally superior. Compare processes against criteria that matter for the chosen material and application: geometry, heat input, achievable microstructure, dimensional control, throughput, post-processing, waste, cost or other relevant constraints. If only technical feasibility is studied, keep business feasibility separate. A laboratory demonstration can show that a process works under defined conditions without proving that the same route is economical or stable at industrial production scale.

Additive manufacturing and 3D printing need complete process context

KTEK0012 covers AM processes, materials, software, equipment, product design and industrial applications. For a thesis, identify the AM process, material or feedstock, geometry, build orientation, key machine parameters and post-processing whenever they can affect the result. If the study redesigns a component for AM, state whether the contribution is geometric, manufacturing, mechanical, economic or several of these. A successful print proves fabrication under stated conditions; it does not automatically prove final-product qualification. When comparing AM with conventional manufacturing, define the comparison basis rather than treating novelty as evidence of superiority.

Separate prototype success from component qualification

A prototype can answer whether a design can be produced, assembled or tested. Qualification asks a different question: whether the component satisfies specified requirements with sufficient repeatability and evidence. If the thesis tests only one or a few specimens, describe the evidence accordingly. Report defects, failed builds and post-processing rather than selecting only the best sample. If industrial scalability is discussed, identify whether it is measured, modelled, supported by company data or only a future implication. This prevents a useful prototype result from becoming an unsupported production claim.

Reverse engineering and industrial metrology require uncertainty

KTEK0031 covers tactile and optical 3D measurement, measurement-system selection, geometric tolerances, surface-finish metrology and measurement uncertainty. A dimensional thesis should report the instrument and procedure together with uncertainty that is meaningful relative to the tolerance or difference being discussed. A difference smaller than measurement capability should not be presented as a strong material or process effect. Reverse-engineered geometry is a reconstruction from measured data. It can reproduce a physical part, but it should not automatically be presented as proof of the original designer’s intent unless other design information supports that interpretation.

Use tolerances and measurement systems consistently

When comparing a CAD model, manufactured part and reverse-engineered model, define the coordinate alignment, datum strategy, filtering and tolerance rules. Changing these rules between samples can create apparent improvements. If several measurement systems are used, explain their different resolution, uncertainty and access limitations. Surface finish, global dimensions and local geometry may require different methods. Statistical quality-control language should be based on enough observations to justify it. For a small thesis dataset, it can be more accurate to report observed variation and uncertainty than to make a broad process-capability claim.

Design optimisation is conditional on the problem formulation

KTEK0030 treats optimisation through design variables, objective functions, constraints, sensitivity analysis and numerical methods. Therefore an “optimal” design is optimal only for the stated formulation. Document what was allowed to change, what was fixed, which objective was minimised or maximised and which constraints limited the solution. If topology optimisation is used, report numerical settings and filters that materially affect geometry. Compare the optimised design with a defined baseline. Do not hide trade-offs by reporting one objective only when other requirements, such as stiffness, mass, manufacturability or durability, are also important to the intended application.

Optimisation is not the same as manufacturability

A numerical optimisation can produce a mathematically attractive geometry that is difficult to manufacture, inspect or qualify. If manufacturability is part of the research question, add manufacturing constraints, process rules or a physical demonstration appropriate to that claim. If it is not evaluated, identify manufacturability as a limitation or future step. The same applies to cost and sustainability. An optimised reduction in mass may be relevant, but it does not by itself establish lower cost or lifecycle impact. Keep computational optimisation, fabrication evidence and product-level implications as separate layers unless the thesis deliberately connects them.

Machine learning for materials science

MTEK0035 focuses on selecting materials-science questions for machine learning, preparing representations, choosing models and evaluating performance with realistic experimental or computational datasets. Define the prediction target before model selection. Document data provenance, sample or material identity, preprocessing, features, split design, model configuration and metrics. A thesis should explain why the split reflects the intended use case. If the goal is to generalise to unseen material families, a random split across closely related measurements may be too easy. The evaluation design should test the kind of generalisation claimed in the conclusion.

Avoid leakage and causal overclaiming in materials ML

Data leakage can occur when information from the test set influences preprocessing, feature selection, hyperparameter choice or training, or when related measurements from the same specimen or batch appear on both sides of a split. Protect the evaluation pipeline and state how independence was maintained. High predictive accuracy is evidence of predictive performance under the evaluation design, not automatic proof of a physical mechanism. Feature importance and Bayesian optimisation outputs can guide interpretation and experiments, but causal claims need additional evidence. Likewise, performance on computational data is not automatically validated experimental performance.

Sustainability and lifecycle claims need a defined boundary

MTEK0025 covers material-selection strategies, environmental impacts across the life cycle, material efficiency, sufficiency, circular economy, sustainability management and policy measures. A thesis should define which lifecycle stages and indicators are actually evaluated. Improved strength, lower mass, recyclability or reduced process waste can be relevant but does not alone prove lower whole-life impact. Distinguish measured environmental data from scenario assumptions. If a full lifecycle assessment is outside the project, use narrower language such as material efficiency, energy demand or end-of-life potential rather than presenting a comprehensive sustainability conclusion that was not calculated.

Company work, confidentiality and NDA boundaries

Company-linked topics are explicitly possible in MTEK0011 and are common in industrial materials. Agree early what data, drawings, process parameters, software, images and results can appear in the examined thesis. Confidentiality should not leave the public or examined manuscript with an unsupported conclusion whose evidence cannot be evaluated. Where sensitive information must be restricted, work with the supervisor and company to design an academically examinable evidence path. Keep company project management separate from the University’s academic assessment. The thesis must still show the student’s scientific reasoning, method and evidence even when the industrial context is confidential.

AI use must remain accountable

University guidance distinguishes responsible AI use from misconduct, and a teacher or supervisor may encourage, restrict or prohibit particular uses depending on learning objectives. Record any material AI use according to current guidance and never outsource responsibility for facts, calculations, code, citations, interpretation or authorship. Verify generated references against real sources. For materials work, be especially careful with generated equations, units, software syntax, material-property values and invented standards. AI can assist a workflow only when the student can inspect, explain and defend the resulting work. If programme, course or supervisor instructions are stricter, those instructions control the thesis practice.

Make figures, tables and data auditable

Every important figure should state what was measured or calculated, units, sample or model condition and the meaning of uncertainty or error bars. Avoid unlabeled screenshots from simulation, microscopy, CAD or machine-learning software when the underlying values can be exported and documented. Keep raw and processed data organised so results can be traced. When an image has been cropped, filtered or contrast-adjusted, preserve the original and describe processing that can affect interpretation. Tables should distinguish measured, simulated, literature and predicted values. This traceability is essential when several tools are combined in one industrial-materials thesis.

A defensible thesis chapter structure

A practical structure is Introduction, Literature Review or Background, Research Questions, Materials and Methods, Results, Discussion, Conclusions and the required front/back matter. Adjust the structure with the supervisor when a computational or company project needs a different architecture. The Methods chapter should allow a knowledgeable reader to understand samples, processing, instruments, software, models, preprocessing and analysis. Results should report observations before the Discussion expands their meaning. The Discussion should compare results with literature, address uncertainty and limitations, and separate demonstrated conclusions from proposed industrial implications. A clean evidence chain matters more than adding unnecessary chapters.

Turnitin checks originality, not technical validity

University guidance makes the Turnitin check mandatory for degree theses and uses it to support review of source use. A similarity percentage is not a technical-quality score. Similarity can arise from references, standard terminology, legitimate quotations or problematic copying, and the report requires human interpretation. The University’s AI-writing indicator is also described as indicative rather than proof of misconduct. Use Turnitin during drafting to correct citation or source-use problems, but do not treat a low percentage as evidence that experiments, simulations, measurements, optimisation or ML results are valid. Methodological validity must be demonstrated separately.

UTUGradu submission, examination and publication

UTUGradu is the University’s electronic process for higher-degree theses. It includes the electronic originality check, examination and approval, electronic publication and electronic archiving. Treat the version entering this process as the controlled final thesis version. Before submission, verify title, abstract, metadata, permissions, confidential-material handling and required files. Do not assume a file accepted by a company system is automatically the correct UTUGradu submission. Follow the current University and faculty instructions at the time of submission because operational details can change even while the 2024-2027 curriculum object remains the same.

A practical thesis timeline

Start by resolving the question, supervisor, evidence level and feasibility. Then complete the seminar research plan, literature mapping and detailed methods before the main experimental or computational work. Reserve enough time for manufacturing, instrument queues, failed samples, reruns, simulation debugging and data cleaning. Analyse results while collecting them instead of leaving all interpretation to the final weeks. Draft Methods and Background early, then Results and Discussion as evidence stabilises. Leave a final period for supervisor feedback, Turnitin, language and figure checks, maturity requirements and UTUGradu. A realistic timeline includes contingency rather than assuming every industrial dependency will work on the first attempt.

Final pre-submission checklist

Before submission, confirm that the manuscript names MTEK0011 as the 30 ECTS thesis and does not merge the separate seminar or Capstone into thesis credit. Recheck that each research question is answered by evidence at the same level, simulation is distinguished from measurement, imaging and metrology claims include suitable sampling or uncertainty, optimisation is not presented as automatic manufacturability, ML evaluation is protected from leakage, and sustainability language matches the lifecycle boundary. Verify citations, data and figure provenance, AI-use compliance, Turnitin, confidentiality arrangements, examiner-facing readability, maturity-test requirements and the current UTUGradu instructions. The strongest thesis is one an examiner can audit from question to conclusion.