The development of Crystal plasticity

Recent experiment have shown the size effect of the materials, when the characteristic length associated with non-uiform plastic deformation is on the scale of micros.The classic plasticity theories can't explain such phenomenon as their constitutive models posses no intrinsic length scale. The new models which contain strain gradient plasticity now are used to explain the experiment. Recently there are also crystal plasticity model to explain and predict the size effect of FCC and BCC.By this time, there have been many models such as CS, SG, MSG ,CMSG.In crystal plasticity ,there have benn models raised by A.Arsenlis, Evers, A.Ma, Gurtin, Yefimov...

So,I want to ask ,are all these model to be the final model in Micro-Materials.If not ,what model can we develop ? What need we do if we want to establish a new model?May be we need to explain these phenomenon from physics and the micro-structure of the material. Maybe ,we need more knowledge of dislocation and other default of materials.But how these deault affect the constitutive models of the Microscale is a big question, How to come through these difficulties.

As I find ,if we have established the new model ,when we calculate the behavior of the material  through the moedl, we  can't use the FE software directly.  we need to use the user defined material/element subroutine of the commercial FE code such as  Abaqus or Marc.

So, I want to ask ,is it difficult to write the code of user subroutine. what kind of help can I meet from the book(list me a booklist or good paper) and websit ,or directly from your sincere advice.

I am a student just gradute as bachelor in china. So please give me some advice.

At the end ,I will be sorry for my poor english expression wasting your time.


Rashid K. Abu Al-Rub's picture

Check these papers

Rashid K. Abu Al-Rub, Ph.D.
Catholic University of America


Rashid K. Abu Al-Rub's picture

Rashid K. Abu Al-Rub,

Rashid K. Abu Al-Rub, Ph.D.
Catholic University of America


Shaoxing Qu's picture

Some references

Young Huang's group have developed both UEL and UMAT within ABAQUS to apply MSG and CMSG theories, respectively. Based on my knowlede, you can develop UMAT if your theory belongs to lower-order strain gradient plasticity, but UEL must be used for higher-order one. CMSG theory is described in the article of International Journal of Platicity V20, p753. UMAT is developed  within the framework of CMSG and a brief introduction can be found in the articles of International Journal of Fracture, V 129, p199, Jornal of Materials Research, V 19, p3423, etc.


How to develop a new?

Thanks


alankar's picture

Alankar Ph.D. Student School

Alankar
Ph.D. Student
School of Mechanical and Materials Engineering
Washington State University, Pullman


alinojumi's picture

UMAT Implementation

Dear Xu Zhang

I have a Problem with ABAQUS, It Doesn't Verify my Fortran

so it cannot call my User Sobroutine for defining a new material

behavior (I'm Trying to define Super-Elastic behavior)

what should I do?

Am I doing somthing wrong or missing any step?

 thanks Alinojumi@gmail.com


Arun Prakash's picture

RE: UMAT implementation

Have you declared your material as a User Material? And have you linked the fortran file that contains your subroutine. If yes, could you be a bit more specific with your problem.


skumaar's picture

ABAQUS UMAT to vary modulus as a function of position in Axisym

S.Kumar,
Solid Mechanics and Materials Engg,
Dept of Engineering Science,
University of Oxford, UK


Arun Prakash's picture

Crystal plasticity UMAT

Prof. Huang developed a crystal plasticity UMAT based on the model of Asaro(1983), Hill and Rice and others. The UMAT is available on public domain. Here is the link

 

http://www.columbia.edu/~jk2079/fem/umat.html

 

You will find documentation and an eg. input file that can be used along with the routine. Note that this UMAT is applicable only to cubic materials and deals only with slip (No Twinning) .


alankar's picture

How to use UMAT for polycrystals

Hi,

I am working on a dislocation density based crystal
plasticity model using FEM. I use UMAT subroutine in ABAQUS. Could anybody help me how you assign initial
orientations to different grains using Euler's angles. Do I need to call
*USER MATERIAL for different grains in your *.inp file. I tried this for a bicrystal, but didnt seem to work.

 

Thanks,

 

Alankar 


Arun Prakash's picture

Props array

alankar:

 Why don't you use the props array (through which you define the properties of the material(s)) to define the initial orientations of the crystals. In the Umat you can easily read in the values from the array and use them for further processing.

 

Arun 


alankar's picture

using PROPS arrays

Thanks Arun,

I am using props arrays. if there is a single crystal, I need to call the *user material once for this single cryatal. i.e. if I have only one element making my crystal, I will call it only once. What about if I have more than one element? I am trying for 8 element body, i.e. each element a different grain having different orientation, making a polycrystal overall. Do I need to call UMAT for each element in my input file?

Alankar

Alankar Ph.D. Materials Science Washington State University, Pullman


Arun Prakash's picture

using FEM??

If I understand it correctly, you use FE and a UMAT for your structure/material. In FE (ABAQUS), every element must be designated a material and the material must defined correctly through the *material card. All you need to do is define eight different materials for your eight different elements.

Hope this clarifies. If not may be you can give me an example input file and I could help you with that.

 

PS: I know it is a pain in the ass to define different materials for different elements. You could probably reduce the effort by using the "*include, input="  card. Decide what material parameters differ from element to element. Group the rest of the parameters in seperate file(s) and include the general files for each *material card  


alankar's picture

re: using FEM

Thanks. Yes I am using UMAT for applying my crystal plasticity model for a fatigue crack problem. I want to use real orientations of grains in my FEM code. That I will do using my OIM experitments results. Yes I think using a separate file for each *material card will be a good idea rather than creating material card in the *.inp file because I might have sometimes mores than 100 grains!!!

Alankar Ph.D. Materials Science Washington State University, Pullman


alankar's picture

re: UMAT

Hi Arun,

Do you think I should create different sections for material sections for each element if I want to define a different property of each material?

Alankar Ph.D. Student School of Mechanical and Materials Engineering Washington State University, Pullman


Arun Prakash's picture

re: UMAT

Alankar:

I am not sure of using sections - it may work, but I am not sure. What I would suggest  is to use element sets,i.e. declare elements belonging to a certain crystal as a set and assign material properties to this set. This not only makes things compact, but also helps in reorganizing things.

 Hope this clarifies your question.

 

Arun 


alankar's picture

Finding Neighbouring Node in ABAQUS

Hi Hemanth,

 Is there any method to determine which node is in next and previous neighbouring position in X, Y, and Z direction in any mesh? I am trying to calculate spatial derivative in my UMAT and I have written a "spider" program for finding neighbouring nodes. However, because of precision problem, it is not perfect. Is there any other method to find neighbouring node or spatial derivatives in UMAT?

 

Thanks, 

Alankar
Ph.D. Student
School of Mechanical and Materials Engineering
Washington State University, Pullman


alankar's picture

Spatial Derivatives in UMAT

Hi All,

Is there any method to determine which node is in next and previous neighbouring position in X, Y, and Z direction in any mesh? I am trying to calculate spatial derivative in my UMAT and I have written a "spider" program for finding neighbouring nodes. However, because of precision problem, it is not perfect. Is there any other method to find neighbouring node or spatial derivatives in UMAT? Can anybody suggest how to determine spatial derivatives in UMAT?

Thanks,

Alankar Ph.D. Student School of Mechanical and Materials Engineering Washington State University, Pullman


Spatial derivative

Nodal coordinate method is the best suited for such purposes.

Vikram Shenoy PhD Consultant, Engineering Integrity Group, Petrofac, Sharjah, UAE.


UMAT subroutine

Dear Arun,

          I am a new user to Abaqus. I am trying to use the subroutine you provided at Columbia website but it doesnt run. It is probably a problem of the Abaqus version... The error msg I am receiving is:

Abaqus Error: Problem during linking - Abaqus/Standard User Subroutines.
  This error may be due to a mismatch in the Abaqus user subroutine arguments.
  These arguments sometimes change from release to release, so user subroutines
  used with a previous release of Abaqus may need to be adjusted.
Abaqus/Analysis exited with errors

The Abaqus version I am using is 6.9. What modifications should I perform? Is there any other reason I get this problem?

 Thank you,

              Panos


Mike Graham's picture

Error message probably deceptive

The UMAT API is very stable and--though I didn't personally check--I highly doubt the message is leading you up the right tree. I have gotten this message before and this was not the actual cause.

Can you explain what process you are using to use this subroutine and what platform you are running on? You might need to install or add-to-the-configuration-file the right compiler and linker. I believe when I ran into this this was the cause.


About UMAT subroutine

Dear all,

         I am currently using UMAt to model two phase material. In order to apply UMAT for polycrystals, as you say this can be done by simply modifying the input file through props array. So everytime UMAT is called the new orientation is used at every point. But how can you model a 2 phase material. How can you say the input file which regions belong to which phase? 

Regards,

      Panos