Discussion Central

Expand all | Collapse all

Multiscale design approach

  • 1.  Multiscale design approach

    Posted 09-17-2019 05:16
    Hello CHE community,

    Traditionally, chemical engineering design has been focused on designing chemical plants to produce well-known compounds of interest in large amounts. Recently, chemical engineers not only need to design chemical plants but also highly valuable chemical products such as biosurfactants or coatings with remarkable characteristics. If we want to prepare chemical engineering students with the ability for designing chemical plants and chemical products (multiscale products), we need to make significant changes in the curriculum. Some efforts have been made to modify traditional courses of transport phenomena and thermodynamics to include molecular properties. However, providing students with more technical content is not enough to assure their understanding of the new concepts. Moreover, it is not clear if the students will directly apply those concepts in the design of chemical products. I would want to ask the opinion of this community about the educational challenges of the multiscale model. Specifically:

     

    What challenges do you think the CHE students need to overcome or can experience when they deal with designing products at different scales?



    ------------------------------
    Ruben Lopez MS
    Research Assistant
    Purdue University
    West Lafayette IN
    ------------------------------


  • 2.  RE: Multiscale design approach

    Posted 09-17-2019 07:16
    Hi,

    I did not understand your definition of "Multiscale products".

    Do you mean, using multi scale models, starting at molecular level, coupling them with models at a larger scale and using them to design plants?

    Can you kindly clarify what you mean?

    Thanks,
    Pavan.

    ---------------------------------
    Pavan Kumar Naraharisetti
    Assistant Professor
    Newcastle University in Singapore
    Singapore
    ---------------------------------





  • 3.  RE: Multiscale design approach

    Posted 09-17-2019 13:38
    Hello,
    Sure, I can clarify a little more. When I talk about multiscale, I am referring to the initial idea of the chemical supply chain that was proposed by Dr. Grossmann and Dr. Westerberg (2000). From the area of process engineering, they proposed a perspective of analyzing the problems of chemical engineering according to their scale: molecules, molecule cluster, particles thin films, single and multiphase systems, process units, plants, sites, and enterprises. Based on that, recent researchers such as Charpentier (2002) and Gani and Ng (2015) propose models to design chemical products based on their scales. For example, Gani and Ng proposed to classify the variety of chemical products in molecules and blends (i.e. refrigerants), devices and functional products (i.e. fuel cells), and formulated products (i.e. paints). I wonder what challenges undergraduate students may experience to learn this approach. Are they prepared enough to design molecules and blends such as chemists? Are the universities providing enough resources to help them understand the chemical and physical phenomena at the level of molecules and multiphase systems? Maybe this approach should be kept for Master and Ph.D. students.

    I would appreciate any comment and if you are interested in learning more about that, I am attaching the described papers.

    Thanks for your time :)


    ------------------------------
    Ruben Lopez MS
    Research Assistant
    Purdue University
    West Lafayette IN
    ------------------------------



  • 4.  RE: Multiscale design approach

    Posted 09-17-2019 19:56
    Hi,

    Allow me to make some simple definitions for convenience.

    Microscale - at molecular level

    At scale - physical properties

    Macro scale - plant level and beyond, (let's exclude supply chain for convenience).

    If I understood correctly, your question is, should we fully integrate and teach UG accordingly ?

    My thinking is as follows :

    The purpose of "microscale" modelling for design of materials is for coming up with new molecules with special properties. Once we do that, we should be in a position to predict the "at scale " physical properties and validate it through experiments. Once we have "at scale" validated, we should be able to operate a chemical plant with confidence.

    In view of this, I think introducing these elements in UG education is important by the way of electives or may be a minor in materials design. If students are interested, they may do a research project. Having said that, this should not come at the expense of core modules of Chemical engineering. Understanding the "at scale" system and "macro scale" design of Chemical plants is the foundation of Chemical engineering.

    Chemical engineering always interfaced with other science and technology subjects, be it environmental engineering, control/instrumentation engineering, industrial engineering, biomolecular engineering or material science. What you are asking is whether the interaction with material science/materials design should be greater. I think this should be left to individual student preference via electives and optional research projects at UG level. When a sufficiently large number of students are involved, we have think about making it core.

    In summary, we should be open to this and do it while measuring the response from students and job market.

    Regards,
    Pavan.



    ---------------------------------
    Pavan Kumar Naraharisetti
    Assistant Professor
    Newcastle University in Singapore
    Singapore
    ---------------------------------





  • 5.  RE: Multiscale design approach

    SENIOR MEMBER
    Posted 09-18-2019 14:38
    As usual, the academic community refuses to recognize the actual state of the industry.  For example, a large majority of Chemical Engineers who follow a non-academic career path, wind up in process operations; the task of making stuff.  Very few wind up in a traditional "design" role.

    Unfortunately, those that do go into design have no concept of how to do Engineering Design.  It is more than being able to run ASPEN or HYSIS (or whatever flavor simulator you happen to use).  It is more than being able to do calculations.  It is focused on the practical aspects of design/construction rather than whatever theory is the current fad.

    It has everything to do with building a safe and reliable plant to produce those "highly valuable chemical products".  As a designer, I want to know the materials of construction (both metallic and soft goods).  I want to know the process limits (Temperature, Pressure, Flow Rate, etc.) .  I want to know reaction rates.  I want to know the potential hazards.

    All of Chemical Engineering distills down to managing Mass and Energy.   Which means being able to accurately measure and control them.

    You should have your students master instrumentation used in industrial settings.  You should teach the proper way to set up an instrument loop.  You should teach practical process control; not the pointless theory taught in academic circles for the past 35 years that is not used in industry (face facts, absolutely noone outside of academia uses Laplace Transforms).  You should teach PHA, HAZOP and LOPA and how that fits into the design cycle.  You should teach Management of Change.  You should teach about local Engineering Standards and Codes.

    These are the useful things that are needed in design, not the dilution you are proposing.

    ------------------------------
    Dr. Keith Dackson , P.E.
    East Aurora NY
    ------------------------------



  • 6.  RE: Multiscale design approach

    Posted 09-18-2019 19:25
    Hi,

    Yes. Still there can be a lot of changes that can be made.

    During my UG, my friends in Metallurgy used to be attached to a steel plant for 6 months to learn things beyond the academia. 6 months from a 4 yr UG is a good amount of time. I did a summer internship at a factory although it was not mandatory.

    Unfortunately, such attachments are uncommon.

    Having said that, we at Newcastle University in Singapore - Singapore Institute of Technology (our UG is a 3yr joint degree) have been fortunate for the support from our industry partners and have been able to send our students for 6 months internship called Integrated Work Study Program.

    Also recognizing the need for practical aspects, we offfer MS in Process Safety and Risk Management at NUiS.

    In addition, the Economic Development Board, Singapore supports new graduates by providing them financial support for up to two if there are Apprentice opportunities in the industry.

    Such academia - industry - government alliance is necessary to address this and train students that are industry ready; not just for Chemical engineering, it applies to many fields.


    Regards,
    Pavan.


    ---------------------------------
    Pavan Kumar Naraharisetti
    Assistant Professor
    Newcastle University in Singapore
    Singapore
    ---------------------------------





  • 7.  RE: Multiscale design approach

    SENIOR MEMBER
    Posted 09-19-2019 10:21
    You are lucky to have the support of industry in your country.

    Here in the US, the perception is that academics prefer to train students to become academics.  The professorate in the US appears to not have any real exposure to "real-world" engineering, whether it is in industry (happens occasionally) or EPC (embarrassingly rare).  They tend to focus on theory, with little to no appreciation for the actual practice of Engineering.  Teaching undergraduate courses is something to be passed to a teaching assistant; who is more interested in doing their graduate research than to waste time with mere undergrads.

    So, going back to the original poster's comments, the students would best be served by getting an education in the fundamentals (I cannot tell you how many times a new BS ChE did not know how to size a pump), not in the arcana of applied research as to how it might somehow impact their job.

    ------------------------------
    Dr. Keith Dackson , P.E.
    East Aurora NY
    ------------------------------



  • 8.  RE: Multiscale design approach

    SENIOR MEMBER
    Posted 09-20-2019 10:02
    "Teaching undergraduate courses is something to be passed to a teaching assistant; who is more interested in doing their graduate research than to waste time with mere undergrads."

    Certainly not my experience at The University of Texas. All of my ChE classes were taught by my professors.

    On the topic itself, there is a serious constraint that has to be respected. An undergraduate program is a four year program, and ChE requires full, heavy loads to do that and properly sequence prerequisites. A successful graduate has an understanding of ChE fundamentals, a broad base and view from a university curriculum, and the training to apply cognitive skills to problem solving. A BS ChE certainly is not a stand-alone professional engineer... yet. To add to the program requires a commensurate subtraction from the program.

    As an engineering manager for a major chemical company, a large part of my job was taking new grads and training them, converting them, to engineers capable of being turned loose on issues of the company. Some had job experiences while in school that helped with that transition, but there is so much to teach. We trained new engineers in the ways we did things, in the support of our business.

    Ours was certainly a traditional ChE application, but many ChEs go a non-traditional route, e.g., what Ruben is describing. There just is not time in a four year program to provide a university education with ChE fundamentals and cover all of the possible specialties such as his. The options are for the company to commit to training and create the growth, or for a BS ChE to extend their schooling toward an advanced degree in the subject. Companies will offer support to faculty to establish this type of graduate degree specialization. If the need becomes broad enough, and the full requirements of a BS ChE are not required, perhaps a new undergraduate program will emerge, e.g. biomedical engineering.


    ------------------------------
    Steve Cutchen
    Investigator
    US Chemical Safety Board
    Houston TX
    ------------------------------



  • 9.  RE: Multiscale design approach

    SENIOR MEMBER
    Posted 09-20-2019 18:57
    Steve:

    Perhaps that was true 30 or more years ago; it was for me.  But it was rare to have a person with industrial experience as a professor.  Even rarer was the individual who did not do R&D in industry.  The attitude of teaching as an afterthought was beginning to arise in the late 1980s, when universities wanted new hires to fund an "eternally funded research program" (as one ad showed its Freudian slip).

    What seems to be missing is how to perform the profession of Engineering.  Why we do what we do, and why in a specific order.  Too few students and professors have ever stepped inside a chemical plant.  Numbers on a page are all well and good, but there is no conception as to how it all fits together.

    Some people rely too much on software, and do not subject the output to any sort of gut check.  I have seen piping layouts look like the old Windows screen saver with multiple bends because they did not interpret the pipe stress analysis correctly.

    My perception right now is that once a person can master the Mass and Energy balances and how to manage them, they can be dropped into any Chemical Engineering subspecialty - this is the company specific training you cite.  Personally, I have worked in Pharmaceuticals, Polyolefins, Electronic Materials, Renewable Fuels, Licensing, and Plant Design, Commissioning and Startup.  My MS was in Acid Rain, and my PhD in modeling 3 D polymeric flows (neither of which served me particularly well throughout my career).

    ------------------------------
    Dr. Keith Dackson , P.E.
    East Aurora NY
    ------------------------------



  • 10.  RE: Multiscale design approach

    SENIOR MEMBER
    Posted 09-21-2019 00:58
    Edited by Steve Cutchen 09-21-2019 01:02
    Thanks, Keith for the thoughtful reply.

    My professors were not industry-types. They were academics. But they were adamant in preparing us for an industrial future. Their goal was to make sure we were prepared in the fundamentals of ChE, but more importantly in the skill of cognitive problem solving. Over and over, this was a fundamental teaching. Sure, we graduated with a firm understanding of material balances, thermodynamics, unit operations, plant design and such... but more importantly we honed our skills in problem solving... the idea that there is not a one correct approach, a one correct optimum.

    My personal growth in real plant operation prior to my degree was in summer jobs that I had in industry. I learned a lot of practical lessons.

    But even with that, I graduated as a green engineer in degree only. Even in a traditional chemical plant environment, it took years for me to be exposed to all of the various parts of the process and to the various techniques required to support them. I'd bet a Dr Pepper that 90% of the work I did was above and beyond what I learned in University. Process Control. Process Hazard Analysis. Incident Investigation. Plant Support. Project support and design. Batch Automation. And that does not even touch the management and supervisory skills I developed. The number of mentors was immense. Both good and bad examples. No one should expect a new BS graduate to hit the ground fully functioning as a professional engineer.

    And that does not even touch the required understanding of how a plant organization operates. How the goals of a Technical , Operations, and Maintenance organization are in a dynamic tension with competing requirements to meet company goals.  This cannot be taught in school. It depends on the industry you choose, the company you choose, the culture of the plant where you work, on on.

    If we had time to share a Scotch in comfy leather chairs, I could hold forth for quite a while on the differences between operations where procedures---work-as-imagined---match work-as-done, and yet other times in the same plant where that does not occur. And still other times where procedures do not, and actually cannot, exist. Learning how to deal with this reality is not the job of a university.

    Ruben mentioned example of non-refinery/chemical homes for ChE such a biopolymers. UTexas offers a degree in biomedical engineering that is probably better suited for that type of industry. Degree opportunities need to continue to grow to meet technology as it expands. But there still needs to be a constraint of a 4 year BS program. And the companies need to always recognize that the new graduates they hire are only schooled in fundamentals. However, and most importantly, as engineering graduates they are also schooled in how to learn and how to problem-solve. Learn how to find the best of these new graduates, target them, hire them. They are your diamonds in the rough. But it is still up to the company and their first-line engineering supervisors to convert an engineering graduate into an actual engineer, to cut that rough diamond rock into a jewel.

    ------------------------------
    Steve Cutchen
    Investigator
    US Chemical Safety Board
    Houston TX
    ------------------------------



  • 11.  RE: Multiscale design approach

    FELLOW
    Posted 09-19-2019 11:07
    Dr. Dackson, you are so correct! Many faculty look "down" on these "trivial" aspects of chemical engineering, without which there would be no chemical industry and all the goods it produces.

    Jack Hipple

    ------------------------------
    [Jack] [Hipple]
    [Principal]
    [TRIZ and Engineering Training Services LLC]
    [Tampa][FL][jwhinnovator@earthlink.net][813-994-9999]
    ------------------------------



  • 12.  RE: Multiscale design approach

    SENIOR MEMBER
    Posted 09-19-2019 12:56
    I completely agree with Dr. Dackson and Mr. Hipple. Undergraduate ChemE education would be well served by the introduction of a process design course that covers the industrial realities of process design. In reality, it would not be a hard course to construct as the very process would be your instrument. Starting from the chemistry of a new product which incorporates the molecular scale and going through the application of the necessary fundamentals for equipment selection and sizing. The incorporation of P&IDs, PHAs, and HAZOPs as the design evolves. Introduction of safety system designs and materials of construction. You could also incorporate the elements of quality control as these tie back to the molecular fundamentals of the chemistry being explored. The difficult part is getting academia to understand the necessity of a course like this.

    ------------------------------
    Wyatt Winkenwerder
    Sr. Pilot Plant Engineer
    Nouryon
    El Dorado AR
    ------------------------------



  • 13.  RE: Multiscale design approach

    Posted 09-19-2019 14:49
    I also agree about the importance of "real" industrial experiences in the education of engineers.  As Pavan said, the alignment between academia, industry, and government is paramount to reach this goal. I can see the issue related to UG education as it may need a more practical approach; however, chemical engineering is so big and requires interdisciplinary work. We cannot deny the diversity of our current students and their different interests; the growing of new industries such as the production of biopolymers; or the different pathways of chemical engineers into the workforce (management, quality, research, policies, etc.). Consequently, the challenge for academia is to provide an education that serves everyone. We need to balance the traditional requirements for chemical plant engineers with the needs of other careers.

    Personally, I believe in an education that helps students not only learn the fundamental concepts but also develop cognitive skills (problem-solving, critical thinking, systems thinking, metacognition) that will allow them to transition in the career that they want. Ideally, we should prepare students to have short learning curves in their first years of working experience. Even if they do not remember exactly how to define the size of a heat exchanger, they should have the preparation to learn that very quickly and apply it to the problem in front of them. With the constant development of new technologies, it is very hard to cover all the details; but what we can do is help students to develop their thinking. I am still thinking if the students need to develop more a particular type of thinking (systems thinking, creativity, critical thinking, etc.) in order to design both, molecules and chemical plants?

    Ruben



    ------------------------------
    Ruben Lopez MS
    Research Assistant
    Purdue University
    West Lafayette IN
    ------------------------------



  • 14.  RE: Multiscale design approach

    SENIOR MEMBER
    Posted 09-20-2019 09:56
    This is an important thread and need a broader perspective in understanding.

    Industry users on the one hand are demanding productivity from their employees. No of days x no. of hours, and each hour of billable rate. This is more dominant in the oil & gas industry, whereupon intellectual know how is a commodity. No doubt software tools and automation have come in. When machines dominate over human intellect, what more can you expect from the society?
    The only knowledge that is expected is of high skills to know particular software and how to extract information out of the tool. More like a skilled driver knowing how to drive an expensive car rather than an automobile engineer knowing about engines and how it functions.

    If the society cannot appreciate the value of human intellect, knowledge and the ability to comprehend and understand the unknown, and get only paid for the skills, then unfortunately the problems will happen. This trend will increase. If business demands only skills like robots, then robots will take over humans.

    Allow the humans to think and do what they are best in doing. Designing robots, but not becoming robots themselves.

    Businesses can't have both the worlds.

    Academics by default are meant to train students to know how to think creatively. In short to know how to design robots.

    Industries bring in the necessary skills through training and broader on the job experiences. And somewhere the roads need to converge.

    Industries expect robots and academics produce robots from factories of "Universities".....does it make sense?

    My views!!!

    ------------------------------
    Eur Ing Santanu Talukdar CEng CSci MIChemE SMAIChE
    Chartered Chemical Engineer
    Independent PAT consultant
    ------------------------------