CemGEMS media¶

Hands-on training #1¶

Q: Is the modified Parrot-Killoh (mP&K) hydration model good for early hydration of cement?

Q: What model can be used for early hydration (to compare with calorimetric curves)?

A: The 5PL (logistic) model could be used for clinker consituents after fitting against the QXRD data.

Q: Can I use the calorimetric data to fit the 5PL (logistic) model?

A: It is difficult to fit 5PL curves against the calorimetry data, which is a sum for all phases reacted.

Q: What is the difference between the process "Hydration-MPK" and "Hydration"?

A: "Hydration-MPK" describes the hydration process using mP&K and/or 5PL models applied to clinker or SCM constituents separately, i.e. it is compatible with "min" type recipes only. 'M' in "MPK" stands for "modified" or "multiple".
A: "Hydration" implements a stepwise reaction extent change applied to the whole Cement and/or SCM material, which is the only choice for the "xrf" type recipes.

Q: Can you see the effect of fineness of Cement or SCMs?

A: Yes, the specific surface area (SSArea) is used e.g. in the mP&K hydration kinetics model. Typical SSArea values for Cement and SCM constituents are provided in template recipes and can be changed using JSON editor (click on small pencil-like icons).

Q: Why there is calcium sulfate among the XRF data?

A: In the "xrf"-type recipe, the Cement composition is set not using constituents, but using the list of chemical formulae. Even though XRF results are reported as oxides (Al2O3, SiO2, ...), other formulae such as "CaSO4" can be used as well. In this case, it is convenient because Ca sulfate is often added to the clinker. The quantity of CaSO4 (as of any formula) can always be set to zero, if needed.

Q: Can we add RH (relative humidity) to study the carbonation?

A: In systems containing aqueous phase after equilibration, RH is close to 100% because it corresponds to the activity of liquid water. We did not test what happens at much lower RH.

Q: How can I calculate a mixture of two different cement types, for example OPC and CAC?

A: The easiest way is to implement in the recipe the CAC chemical composition as a formula list in a constituent in SCM material, and enter its quantity. Use the JSON editor to duplicate formulae or constituents, if needed.

Q: Is it possible to have two types of a mineral with different chemical formula (eg. belite with different ratio of Ca/Si/Ti/..)?

A: Yes, this is possible. You can duplicate the Belite in the JSON editor and change composition of the duplicated one. Better to move the second Belite to the end of constituents list, to avoid breaking the hydration process simulations.

Q: Can we extract saturation index values (and its variations) of different phases during the process simulations?

A: Please, use GEM-Selektor for this purpose (CemGEMS only displays in the Equilibrated part the stable phases that all have saturation index very close to 1 (or 0 in log scale)).

Q: I do not see the volume of pores in the result, am I wrong?

A: The volume of pores is the same as the volume of aqueous solution - the first phase in Equilibrated part of the recipe.

Q: Because the Debye-Hueckel equation cannot account for high ionic strength, then pore solution results are only qualitative?

A: The extended Debye-Hueckel equation is valid to up 1-2 molal, whereas pore solution in most cases has ionic strength around 0.5 molal, well within the validty of the model. Exception: salt attack, where the ionic strength goes above 2 molal upon addition of a few gram salt per 100 g cement at w/b less than 1.0.

Q: Is there any option for adding Pitzer model?

A: Yes, we anticipate adding SIT and Pitzer models as alternative GEMS3K initialization documents in future.

Q: Phosphate phases are included with the usual Portland cement phases. Is the merged database available somewhere?

A: Yes, we did merge Cemdata18 database with phosphate data and Na and Ca-zeolite database, all recently published and downloadable from https://www.empa.ch/cemdata.

Q: After the CemGEMS upgrade has been announced, I still do not see new options and templates. What should I do?

Q: What happens to my data after the upgrade and a new log-in?

A: Your previous recipes and processes are safe, but may need to be re-calculated.

Q: What about the confidentiality of data submitted to your servers?

A: Technically, the user data (recipe, process and plot documents) are saved in the database on the server and sent from/to the client browser via the encrypted connection. Only the root who maintains the server can access all the content in the database. The root can see user password only in encrypted form, unusable at login. The users cannot see and cannot access each others data.

Q: Is it possible to send feedbacks if someone does not use a mail client but rather webmails?

A: Yes, just use email addresses "issues@cemgems.org" for reporting issues and bugs; "comments@cemgems.org" for comments, and "improve@cemgems.org" for suggestions.

Q: Is it possible to change the axis titles of the plots?

A: Currently, it is possible to edit main title of the plot use JSON editor and editing the field "titles" in the current Plot document. Clone "$default" plot document if you would like to preserve the changed title. In future releases, it will be possible to edit also axis titles.
A: Alternatively, export the plot data into csv format and replot it in excel or another software.

In future releases, the step in process definions should be set to auto by default. Then the users will not need to change the number of steps when modifying the interval and stepsize.
The idea of using templates is that one can do meaningful calculations with just a few clicks. Of course each user should make changes for his specific composition/recipe.
CemGEMS developers are writing an article for RILEM letters, which will describe the main processes and options, and give a comparison with GEM-Selektor, which might be of additional help in the future.
The system is even converging when the amount of water is under stoichiometric. Yes, GEM converges sometimes, but in reality the reaction of the clinker will stop when all the water is consumed into hydrates.