De seneste årene er det oppdaget slitasjeskade på offshorevindturbinenes fortøyningsplass, hvor båter kommer inntil for på og avlastning av drift og vedlikeholdspersonell. Hensikten med denne oppgaven er å se på beleggenes egenskaper i forhold til slitasje og å få en forståelse av hvorfor slitasjeskade oppstår ved fortøyning mellom båt og vindturbin. Beleggene som er valgt ut til testing, er beleggsystemer som i dag brukes på offshorerelaterte strukturer, som oljeplattformer og vindturbiner De mekaniske egenskapene til beleggene ble undersøkt ved hjelp av vickers mikrohardhetsmåler, ruhetsmåler, slitasjeapparat og strekkapparat. • Polyester inneholder glasspartikler og fikk påvist høyest massetap, glasspartiklene kan ha falt av under slitasje og bidratt til skade på beleggoverflaten. • I slitasjetesten ble det funnet at gummi heftet på beleggenes overflate. Den avsatte gummien kan derfor ha virket som et beskyttende lag. • Det ble funnet at lang herdetid øker slitasjeegenskapene til belegget. • For epoksybeleggene og varmforsinket stål var slagfastheten bedre enn for de andre beleggene, dette skyldes god heft mellom lagene. Testene i denne oppgaven viser at det er ulike årsaker til at skade oppstår i belegg på grunn av fender. Det er funnet at glasspartikler, høy ruhet, dårlig heft mellom strøkene og kort herdetid kan ha vært årsaken til skade på belegg på grunn av fender. Et Epoksy belegg er funnet til å være det som vil være best egnet til å brukes på fender- rør.

In this master thesis experiments were conducted to determine the abrasion-corrosion properties of a steel designed for TBM tunnelling. This was done by three different tests, reciprocal ball-on-plate, rubber wheel and hyperbaric soil abrasion test. The reciprocal tests were done by rubbing steel balls onto rock obtained from a tunnel boring site in the Faroe Islands. The test were performed in dry conditions, in water from the same site as rock and a mixture of the water and a foam designed for use in hard rock tunnel boring. During these tests no measurable weight loss was produced, but the water and the foam caused some pitting to occur on the steel surface. The water and foam also provided some lubrication, where the foam clearly was the best lubricant by producing elastohydrodynamic lubrication (EHL). The rubber wheel tests were performed in different combinations of water, sand, oil and a foam specially designed for soft ground tunnel boring. The sands used were sand used for making casting moulds and sand obtained from a soft ground tunnel boring site in Israel. The two different water samples used were obtained from the site in the Faroe Islands mentioned above and the previous mentioned Israel site. The tests showed that adding foam to the sand and water mixture significantly reduced the measured weight loss. However, as for the reciprocal ball-on-plate test, the foam caused pitting to occur. The same corrosion effect was observed for the oil additive, but the oil did not provide enough lubrication to avoid abrasive wear. Consequently, the test with the oil additive produced the biggest weight loss of all the rubber wheel tests. The hyperbaric soil abrasion tests were performed in the sand obtained at the site in Israel. The sand was tested both dry and saturated with water from Israel. Tests showed that the measured weight loss from these two tests was similar. However, SEM pictures revealed that in the test with sand saturated with water, both abrasion and corrosion had occurred. While in the dry sand, only abrasion had occurred. This indicates that in this test the hardness of the steel and abrasiveness of the sand is more decisive for the weight loss than the corrosivity of the solution.

When a liquid SiMn alloy is cooled, carbon is dissolved from the melt. The dissolved carbon either forms graphite or reacts with silicon from the melt to form SiC. The goal of this thesis has been to determine how temperature fluctuations affect the formation of SiC and graphite in the SiMn process. The focus of this thesis has been on dissolved carbon from the SiMn alloy due to cooling of the system, and the formation of graphite and SiC from this dissolution of carbon.The experiments conducted in this thesis is done with near constant silicon content, and temperature as the main variable in addition to carbon. The investigation was carried out by heating SiMn alloys several times, forming SiC and graphite. Carbon from the crucibles dissolved into the liquid SiMn alloy to reach equilibrium between carbon and silicon. This is equivalent to an industrial furnace where coke can dissolve carbon into the SiMn alloy, and thus form SiC or graphite. Accumulation of graphite and SiC is believed to be bad for furnace operations.SiC that forms in a Mn-Si-Fe-C saturated alloy can have an interface structure of granular, angular grains and as dendrites. When SiC accumulates in the Mn-Si-Fe-C system, clusters of SiC particles with intermediate SiMn metal can form. This study shows that SiC clusters is typically a decade larger than the average SiC particle. The SiC particles settles at the top of the melt when a cooling rate of 18°C per minute is used.Graphite that forms in the Mn-Si-Fe-C system have an interface structure of flakes. When the carbon content of the liquid SiMn is high enough, nucleation of new graphite grains happens instead of grain growth. The graphite flakes did not settle, but remained evenly dispersed in the SiMn alloy. This may be due to high viscosity in the melt, high friction on the graphite flakes or the buoyancy can be fairly equal to the drag of the graphite particles.The rate of formation of SiC and graphite in Mn-Si-Fe-C saturated alloys is reduced slightly by each consecutive temperature cycle. This suggest that SiC and graphite formed during the first cycles only partly dissolve into the liquid SiMn alloy to reach equilibrium in the system.

Since the late nineties weldable supermartensitic stainless steels (SMSS) has been used in subsea flowlines, and more than 400 km is installed. Hydrogen related problems in the weld fabrication stage or hydrogen induced stress cracking (HISC) due to hydrogen embrittlement under cathodic protection have been reported and testing to better understand the behavior of these materials is of interest. The present work is focusing on the effect of double heat cycles on carbide precipitation and secondary hardening in two high grade SMSS, one alloyed with Ti and one not (0.007% Ti). The experimental work included heat treatment of both steels in single and double cycles in both oven and in induction heating weld simulator. All specimens were heated to 1000 degrees celsius for 30 min to austenize the material and air cooled before simulating second heat cycles of 1-60 min in the temperature range 500-600 degrees celsius. The microstructures resulting from the heat treatments were characterized by standard metallographic techniques, hardness testing and transmission electron microscopy (TEM). The results from hardness testing show a secondary hardening effect in both steels for all temperatures tested, with a maximum for samples treated at 550 degrees celsius. For all samples there was also a significant drop in hardness from the hardness of the single cycle sample to the 1 min tempered double cycle samples. From investigation in TEM it was found that, in the steel alloyed with titanium TiC precipitate during tempering and in the steel not alloyed with titanium Cr23C6 precipitate during tempering at 550. The difference in secondary hardening found is explained by the difference in carbide precipitation mechanism. In the steel alloyed with titanium a higher secondary hardening effect was found. No carbide precipitation was found in the single cycle heat treated state of the steels.

Det har i dette arbeidet blitt undersøkt hvilke effekter relativt store mengder silisium har på egenskaper i stål. To ulike deler i stål, tinder og fjærer, som jordbruksselskapet Kverneland Group ASA produserer, har blitt undersøkt nærmere. Felles for tinder og fjærer er at de er utsatt for store mekaniske belastninger og trenger derfor høy styrke. Begge må tåle sykliske belastninger og de krever derfor stor utmattingsmotstand slik at de ikke går til brudd for tidlig.Tindelegeringen som ble studert var 33S17B og de fire fjærlegeringene var 46Si7, 55Si7, 50CrV4 og 35M13B. Felles for tinder og fjærer var at både tindelegeringen, 33S17B, og to av fjærlegeringene, 46Si7 og 55Si7, hadde et høyt silisiuminnhold på henholdsvis 1,6 %, 1,6 % og 2,0 wt.%.Hovedmålet med denne oppgaven har vært å finne ut hvilken effekt det relativt høye silisiuminnholdet i 33S17B, 46Si7 og 55Si7 hadde. For 33S17B var det interessant å se i hvilken grad silisium påvirket senterseigringene i godset. For fjærlegeringene ble de med høyt silisiuminnhold, 46Si7 og 55Si7, sammenlignet med legeringsalternativene med langt lavere silisiuminnhold, 50CrV4 og 35M13B. I løpet av dette arbeidet har kjemiske analyser og varmbehandling av tindelegeringen 33S17B blitt utført. De varmebehandlede mikrostrukturene har blitt studert i lysmikroskop og mikrohardhetsmålinger ble utført. Videre ble de fire fjærlegeringene 46Si7, 55Si7, 50CrV4 og 35M13B varmebehandlet og anløpt på ulike måter. Prøver av legeringene ble studert i lysmikroskop, hardhetsmålinger ble gjort på både mikro- og makronivå og strekkforsøk ble utført. Det høye silisiuminnholdet i 33S17B, 46Si7 og 55Si7 førte til at ferritt var en stabil fase ved relativt høye glødetemperaturer. Ved et lavere silisiuminnhold ville ikke ferritt, men kun austenitt, vært stabil for de samme glødetemperaturene.I 33S17B forekom blant annet silisium som senterseigringer i leveringsmaterialet. Konsentrasjonsforskjellen mellom seigringen og legeringens gjennomsnitt var på omtrent 1 wt.%. Dette førte til at seigringene inneholdt forekomester av ferritt etter bråkjøling fra så høye temperaturer som minst 1000 °C. Varmebehandling ved 1050 °C gjorde at seigringsferritten ble austenittisert og omdannet til martensitt ved bråkjøling.For fjærlegeringene 46Si7 og 55Si7 førte det høye silisiuminnholdet til relativt kraftig avkulling ved en herdetemperatur på 900 °C. Ettersom overflaten ble avkullet i løpet av varmebehandlingen gjorde det høye silisiuminnholdet at austenitten ikke lenger var stabil ved glødetemperaturen. Den karbonfattige austenitten i overflaten ble omvandlet til ferritt ved glødetemperaturen. Etter bråkjøling forble ferritten uforandret.Som for legering 33S17B førte også en høyere herdetemperatur til at 46Si7 og 55Si7 ble fullstendig austenittisert i forkant av bråkjølingen. Det ble funnet at 46Si7 og 55Si7 bør bråkjøles fra henholdsvis 950 °C og 1000 °C for å unngå et ferrittsjikt i overflaten.Legeringene 50CrV4 og 35M13B hadde et langt lavere silisiuminnhold og hadde derfor ikke et ferrittsjikt på overflaten. Ved å bråkjøle 50CrV4 fra 900 °C oppnådde legeringen den høyeste overflatehardheten blant samtlige avkullingsprøver. Dette viste at 50CrV4 er legeringen med den antatt høyeste utmattingsmotstanden. Det ble imidlertidig påvist mulig korngrenseoksidasjon i overflaten på 50CrV4 etter noen av varmebehandlingene, noe som i så fall vil svekke utmattingsmotstanden.Legering 35M13B bør austenittiseres og bråkjøles fra en temperatur ikke høyere enn 900 °C for å oppnå en høyest mulig hardhet på overflaten. Strekkforsøkene viste at 35M13B var legeringen med størst arbeidsherding.

Laminatkompositter av plater av AA3103 aluminium + copper, samt AA3103 + C24000 messing, har blitt laget ved å anvende en prosess for akkumulert valsesveising (ARB). To forskjellige kobber/aluminium-stabler ble utsatt for 6 sykluser ARB ved en valsetemperatur på 350°C, og beholdt en kontinuerlig lagstruktur gjennom hele prosessen. Den relativt lave graden av kaldbearbeiding gjorde imidlertid at styrken økte lite eller ingenting med økende antall passeringer. 3 andre paralleller med originalstabler ble lagd av kobber, ikke-anløpt messing, og anløpt messing, der hver plate ble stablet mellom to aluminiumsplater som var betraktelig kaldvalset. Den første passeringen ARB foregikk ved 350°C, mens de påfølgende passeringene skjedde ved romtemperatur (CARB). Hver prøve som ble laget i løpet av denne prosessen ble kritisk oppsprukket, langs midten av plata, langsmed valseretningen, i løpet av den 5. passeringen ARB. I tilleg så hadde store, akkumulerte tøyninger forårsaket betydelig sprekkvekst langs kantene av laminatplatene.Etter passeringene før kritisk sprekkvekst oppstod, viste de kaldvalsede prøvene økende styrke med økende antall passeringer gjennom valsen, helt til det punktet hvor kobber/messing-lagene hadde sprukket opp og ble diskontinuerlige. De oppsprukne lagene bidro lite til styrke i både strekk- og bøyetester, samt at disse materialene gav prøver med innbyrdes større avvik mellom prøver tatt fra forskjellige steder på samme plate. Tidligere forskningsarbeid har funnet at forskjeller i plastisk flyt kan forårsake instabiliteter ved grenseflatene, og at dette fører til necking av lagene til det hardeste materialet. I løpet av arbeidet utført i denne rapporten, ble forholdet mellom lagtykkelsen til de ulike materialene balansert for å motarbeide denne effekten. I tillegg så viste de fremstilte prøvene en hittil oversett instabilitetsmekanisme: skjærdeformasjon. Skjærbånd ble observert i kobberlagene under EBSD-undersøkelser, og antatt korresponderende bølgelignende deformasjonsmønstre, som forplantet seg på tvers av lag, ble observert i lysmikroskop.Videre ARB-passeringer med de utforskede materialene vil ikke være mulig uten noen form for varmebehandling eller anløping mellom enkelte passeringer, siden det vil bli nødvendig å løse opp spenninger i de svært arbeidsherdede materialene.

A characterization of the softening behaviour of four different commercially pure aluminium alloys has been carried out. The work is related to the MOREAL project (Modelling towards value-added recycling friendly aluminium alloys), where the main goal is to quantify the effect of the elements in recyclable aluminium alloys on microstructure and mechanical properties during thermo-mechanical processing. Typical elements are iron (Fe), silicon (Si) and manganese (Mn), and the alloys studied in this work contain Fe and Si with different amount and ratio: alloy A1 with 0.15 wt% Fe and 0.05 wt% Si; alloy A2 with 0.15 wt% Fe and 0.15 wt% Si; alloy B1 with 0.5 wt% Fe and 0.05 wt% Si and alloy B2 with 0.5 wt% Fe and 0.15 wt% Si.The as-cast material of all four alloys were homogenized at 600 °C for 24 hours followed by a cooling sequence to 450 °C, implying a total dwell time of 160 hours. The alloys were then cold rolled to a strain of 2.6 and isothermally annealed at temperatures from 275 °C to 375 °C, and the physical and mechanical properties were followed with electrical conductivity and hardness measurements. The microstructure and texture has been investigated with electron backscattering diffraction (EBSD) in scanning electron microscope (SEM), optical light microscopy and orientation distribution functions (ODF) from X-ray diffraction. Alloy A1 and A2 where fully recrystallzed after 10 000 seconds when isothermally annealed at 300 °C. Alloy B1 and B2 are slightly faster to reach the fully recrystallized state than alloy A1 and A2. The decrease in mechanical properties during softening was nearly linear on a logarithmic time scale, especially for alloy B1 and B2, with the onset of recrystallization difficult to seperate from the recovery. Electrical conductivity measurements showed that there was minimal concurrent precipitation. Images of the microstructure of the samples annealed at 275 °C show a long recovery phase followed by recrystallization. Particle stimulated nucleation (PSN) sites seems to be an important nucleation mechanism as it is found that the initial grain size have little effect on the softening kinetics, indicating that nucleation on old grain boundaries is of little importance. The recrystallization texture is weak with the typical Cube orientation slightly rotated around the normal direction. The recrystallized grain size was found to be smaller in alloy B1 and B2 (16-20 μm) than in alloy A1 and A2 (21-27 μm), most likely due to more PSN sites in these alloys. The grain growth that followed after recrystallization was found to be slightly larger for alloy A1 and A2 than alloy B1 and B2, most likely due to the effect from solute drag.

Intragranular chromium nitrides is a phenomenon with detrimental effects on material properties in superduplex stainless steels which have not received much attention. Precipitation of nitrides occurs when the ferritic phase becomes supersaturated with nitrogen and there is insufficient time during cooling for diffusion of nitrogen into austenite. Heat treatment was carried out at between 1060◦C and 1160◦C to study the materials susceptibility to nitride precipitation with varying heat treatment temperature, nitrogen content and microstructure. Microhardness tests and Charpy v-notch tests were also carried out to investigate the nitrides effect on material properties. No standard method of quantification of nitrides exist. A method of quantifying the precipitation based on area fraction of nitrides was therefore introduced. The results show an increased amount of nitride precipitation with increased heat treatment temperature. A coarse microstructure with a large austenite spacing was found to promote higher fractions of nitride precipitation while nitrogen content was found to affect the amount of precipitation in less extent. The intragranular nitrides cause precipitation hardening in the ferrite and the precipitation was found to be at its most severe in the center of the ferritic regions, with precipitation free zones close to the phase boundaries. The microhardness of the phase was affected accordingly, with increasing hardness towards the center. Charpy v-notch test results show that nitride precipitation causes an embrittlement of the steel while intragranular secondary austenite improves the impact toughness of the material as it shortens the dif- fusion distance of nitrogen, decreasing the materials susceptibility to intragranular nitride precipitation.

ALM-based production of Titanium components shows great promise in supplying cost-effective products within industries such as energy, chemical processing and aerospace. In common for all is high quality demands. This necessitates extensive research into the intricacies of this production process. Complex microstructures comprised of columnar remnants of directionally solidified β-Titanium, with interior inhabited by colonies of finer α-plate structures were found in samples produced by layered plasma welding of alloy Ti-6Al-4V. Utilising the powerful characterisation tool of in-situ tensile tests combined with EBSD analysis allowed qualitative correlation between microstructural features and deformation characteristics. Non-uniform deformation occurs due to a strong variation of strain response between colonies. Prismatic and basal slip systems are active, with the prismatic systems contributing to the most severe deformation through coarse, widely spaced slip bands. Certain colonies behave as microstructural units, with easy slip transmission across the entire colony. Other colonies show significant deformation mismatch, with local build up of strain gradients and stress concentration. Segmentation occurs due to the growth morphology imposed by the columnar grains. Tensile tests perpendicular to columnar structures reveal strong deformation localization. Connections are made between the peculiarities of the production process and the observed microstructure and deformation mechanisms.

Arctic steels, like the API X70 steel explored in this master thesis, are developed to withstand very low temperatures, with a ductile/brittle transition temperature below −60◦C. However, during welding, brittle phases like martensite-austenite (M-A) islands may form in the heat affected zone (HAZ). In this master thesis in situ heat treatment and quenching experiments have been combined with electron backscatter diffraction (EBSD) inside a field emission scanning electron microscope in order to simulate a second weld cycle simulation. In conjunction with this a quenching device has been designed with respect to an existing hot stage.Inside the SEM, it was found difficult to recreate the characteristic microstructure of API X70 steel from a traditional second cycle thermal weld simulation. This was probably due to the slow heating rate and the excessive high temperature holding time. The developed quenching device was based on gas quenching and produced a steel sample cooling time from 600◦C to 400◦C, ∆t6/4, of approximately 6.5 seconds in every experiment. As a result of the gas quenching substantial amounts of oxide contamination, covered the sample surface following the experiments. This contamination layer produced artifacts, such as unreasonable amounts of austenite in the EBSD phase maps.Despite the quenching, islands of austenite (FCC) observed at elevated temperatures transformed into a low strain BCC structure. Some austenite islands showed a memory- effect during phase transformation, while others adapted the orientation of the surround- ing structure. No retained austenite was observed in the API X70 steel after the in situ gas quenching, probably due to insufficient cooling rates.