Final Project Commit For Adam Read

[aread85]

This has the note book containing my final project.

[labarba]

You forgot to put your byline, ideally with a copyright notice. We encourage you to put a CC-BY license on your work, so it can be shared, reused, remixed (with attribution).

Also, don't you have a couple of references to cite?

[labarba]

Also, in your final plot, it would be helpful to have different colors or line styles and a legend!

[aread85]

Added CC-BY, references, and legend

[labarba]

The quote "All models are wrong, some are useful" is credited to George Box, a British industrial statistician http://en.wikiquote.org/wiki/George_E._P._Box

I feel that the Introduction could've used a lead-in to the subject of the notebook, since it starts with a general roundabout on modeling and the next section jumps straight into CSTRs. (By the way, the apostrophe is not necessary when you add an "s" to make plural an acronym.)

"The different times that the atoms spend in the reactor means that there is a residence time distribution RTD. The RTD is an important design feature for increasing the efficiency of the reactor." —> Since many of your readers may not be familiar with the topic, I feel that the explanation here needed to be more detailed. You could also reference the textbook (Fogler) here. For example, something like: "Fogler (2006, p.870) gives the background on the concept of residence-time distribution (RTD). Because most chemical reactors are non-ideal, all the atoms of material stay in the reactor a different length of time, i.e., have different residence times. This leads to the idea that there is a distribution of residence times. The residence-time distribution (RTD) of a given reactor characterizes the quality of the mixing and thus is an important attribute." (Feel free to reuse this bit of text, which I just wrote after reading Fogler's discussion in §13.1.1).

Your following passage, explaining the pulse input experiment, follows Fogler quite closely (at times, paraphrasing him), and thus you must mention this, crediting your source properly.

In your first equation, it would be better if you used $V_t$ to get a subscript; otherwise, it looks like the multiplication of V and t.

And to get numbers in your equations (when using the numericalmoocstyle.css style sheet), you need to enclose the equation with the commands \begin{equation} … \end{equation}

In your equation for RTD, you need to write the limits of integration for C, don't you?

I, personally, am not familiar with the theory and practice of chemical reactors. I tried going back to your source to see if I could better understand what you are demonstrating in this notebook. But still I remain somewhat confused. You combine the equation for concentration and the time rate of change of tracer into an ODE for tracer mass: dm/dt = m v/Vt (which should be negative because the tracer mass is decreasing). You used Euler's method to integrate this simple ODE with given initial conditions. With the solution, you get concentration over time, and numerically integrate this with SciPy to get E(t). It's a nice exercise, but E(t) for a perfectly mixed CSTR has an analytical solution (Fogler, p.887), so what is the point of doing it numerically? Also, the analytical solution is E(t) = e^{-t/tau} / tau ... this function should have the value 1.0 at t=0 ... I suppose that is normalized? Why does your result show a value of ~0.11 at t=0 ? (again: I'm not familiar with this field)

In the passage "Based on the model, it looks like 100 grams of tracer … " I think you are saying that by t = 30 or 40 min the concentration is close to zero, so you conclude that a field test should take samples for that long. The E curve you show is for a perfectly mixed CSTR, and you are using that to guide an potential test on a non-ideal reactor?

As you can see, I was confused by your presentation, despite good effort on my part —going back to Fogler and reading through Ch. 13 and all. BTW, I read it online on this location: http://www.umich.edu/~essen/html/byconcept/cdchap/13chap/ECRE_CD-CH13.pdf

The second part where you study CSTRs in series is interesting, but I was still unsure of the aim of the series model. The point seems to be that the model of ideal reactors in series could be used to guide the design of testing on real reactor systems. I hope I got that right.

Typos engineer's —> engineers (again below) cuases —> causes vessesl —> vessel a process know —> a process known skimed —> skimmed indefinaetly —> indefinitely reveiw —> review agian —> again (there is a period in this passage where there should be comma) it's self —> itself