Test Structure And Methods For Fracture Strength Of MEMS Thin Films

Various test techniques have been developed to determine the mechanical properties of microscale components such as nanoindentation,bulge,torsion and bending etc.Most of these methods are complex in nature and require detailed analysis.Furthermore,an external or off-chip actuation method is required to apply stress on the test specimens which can be a source of error if the specimen and load source are not well aligned.The aim of the thesis is to design a test structure and demonstrate the fracture strength of MEMS thin films.In-situ chevron-shaped thermal actuators?TAs?can create a full stress-strain curve for each sample of interest because they can provide large force and displacement.A voltage is applied across the beams of the actuator that results in joule heating and expansion of the beams.This leads to rectilinear displacement in the connected components.The disadvantage of this technique is the large heat flow and high temperatures that the specimen may experience during testing,which can alter the mechanical properties of some materials.To deal with this problem,a heat sink is added to the structure to limit the heat flow to the test specimen.Chevron thermal actuator with 6 different design dimensions is designed and simulated in COMSOL Multiphysics and the results are analyzed.The shape of the actuator remains the same.Only the length and the number of beams,length and width of the shuttle and the pre-bent angle are changed in each model.In the 1^st model,the length of the actuator beams is 116?m,width is 10?m and the thickness is 10?m,the length,width and thickness of the shuttle as 300?m,10?m,10?m and the pre-bent angle of the beams is 10o.The length of the heat sink is 200?m,width is 15?m,and thickness is 10?m.The length and width of the test specimen is fixed due to the design constraint of the PiezoMUMPs.The length of the test specimen is 10?m,width is also 10?m and thickness is 0.7?m.The applied voltage across the beams is 5 volts.When the design is simulated in COMSOL Multiphysics,it produced a maximum stress of 1.6 GPa on the test specimen.The maximum temperature of the thermal actuator is 940oC and the test specimen is 62oC.In the 2^nd model,the dimensions of the device remain the same as the 1^st model.Only the pre-bent angle of the beams is changed.The pre-bent angle is now 15o.When the model is simulated in COMSOL Multiphysics,it produced a maximum stress of 1.4 GPa on the test specimen.The maximum temperature of the actuator and the test specimen remains the same as the 1^st model.Both of these models heat up above 900oC.That's why they are not compatible to be used.In the 3^rd model,the dimensions of the device are changed.The length of the beams and the heat sink is changed.Also the width of the shuttle is changed.The length of the actuator beams is now 200?m,the shuttle width is 20?m.The length of the heat sink is 300?m.The pre-bent angle of the beams is 15o.When the model is simulated then it produced a maximum stress of 2.8 GPa on the test specimen.After changing the length of the beams,the maximum temperature of the actuator dropped to 630oC.The maximum temperature of the test specimen is 67oC.In the 4^th model,all the parameters of the actuator are kept the same as the 3^rd model.Only the pre-bent angle of the beams is changed.The pre-bent angle is now 10o.When the model is simulated in COMSOL Multiphysics then it produced a maximum stress of 3.13 GPa on the test specimen.The temperature of the actuator and the test specimen remains the same as the3^rd model.The maximum temperature of the 3^rd and 4^th model is still high i-e 630oC.In the 5^th model,the number of beams is increased while the dimensions of the model remain the same as the 3^rd and 4^th model.The number of beams is now 10 and the applied voltage is 4V.The pre-bent angle for this model is 15o.When it is simulated in COMSOL Multiphysics then it produced a maximum stress of 2.13 GPa on the test specimen.The maximum temperature of the actuator dropped to 430oC.The maximum temperature that the specimen bears is 68oC.As it can be seen that the temperature of the actuator dropped but at the same time the maximum stress is also reduced.In the 6^th model,only the pre-bent angle is changed and the rest of the dimensions remain the same as the 5^th model.The pre-bent angle is now 10o.The temperature of this model is the same as 5^th model.The change in the pre-bent angle to 10oimproved the stress value that it applied on the test specimen.The maximum stress that it produced on the test specimen is2.56 GPa.After executing all the 6 models of the chevron-shaped actuator in the COMSOL Multiphysics,it has been found that the 6^th model provides large force and displacement as compared to the other models at low temperature.Also the optimum pre-bent angle of the actuator beams for creating large force and displacement has been found to be 10o.