5 replies to this topic

[Jiten_process]

Dear All,

 

I need your guidance in the subject matter. We are working on one proposal of crude oil processing facility. we have FEED data with us and in that well fluid compositions are given. crude is relatively heavy crude with API density ranges 16° - 19°. In this case, client has not provided crude oil ASSAY in terms of D86 or TBP data, instead given component wise mole fraction including C31 to C36+ hypo components property to generate in hysys. Attached below is the input data that we have received from client.

 

Since I am new in characterizing HEAVY oil in hysys, i am not sure whether the data given in attached spec for hypo component are sufficient to generate accurate hypo C31 to C36+ component. The reason I am asking is, when i genrated hypo based on the data given in attachment and generated the stream in hysys inputing all the individual component composition. I am getting very weird result in hysys. I am getting density below 1 at specified pressure and temperature. 

 

my questions mainly are

 

1) Is data provided as attached is sufficient to model crude oil accurately in hysys. if yes, where am i goofing ?     I have used peng robinson thermo model.

2) Is there anything special consideration to be given to characterize HEAVY oil in hysys. The crude which is in question in our case is having API gravity range as 16-19°. crude is sweet (no H2S). 

Attached Files

•  crude oil data.JPG   124.12KB   74 downloads

[PingPong]

I am sorry to say that those data are of little use.

 

The C10 through C35 components are all specified to be alkanes, which is nonsens.

For example: C35 is named as pentatriacontanes, which are isomers of C₃₅H₇₂ (all with a molecular weight of 437).

Even if all C35 molecules in the crude would be C₃₅H₇₂ isomers (which they are not) they would not all have the same specific gravity or the same atmospheric boiling point: there would be a large amount of isomers with different boiling points and different specific gravities.

In reality the C35 molecules in crude are not only alkanes but mainly aromatics and polyaromatics, of which you know nothing because that is not included in this analysis.

 

The same applies to all fractions C10 through C34, they are in reality not pure alkanes either, but a mixture of alkanes, naphthenes, aromatics, polyaromatics and polar polyfunctional compounds (polyaromatics containing O and N atoms).

 

And for none of the fractions in the analysis the average atmospheric boiling point is given.

See attached figure for components normally present in crudes: it shows that there is an enormous difference in atmospheric boiling points of fractions with the same number of C-atoms. For example: for 24 C-atoms there are molecules with boiling points ranging from 390 ^oC (nC24) to ~700 ^oC (polar polyfunctional). For 35 C-atoms it ranges from 490 ^oC to ~800 ^oC.

 

Note also that the C36+ fraction constitutes the bulk of the weight and volume of the crude, but it is only given as one fraction in the analysis. So much heavies is not strange for a very heavy crude oil like yours with an API gravity of only 16 - 19. Unfortunately that kind of heavy crudes usually contain a lot of aromatics and polyaromatics of which you know nothing as they are not included in the analysis.

 

EDIT: I suggest you find out the name of the crude, and what ASTM (or other) method was used to generate the above analysis of the crude sample.

Attached Files

•  Atm Boiling Point versus Carbon Number of Crude Oil Molecules.jpg   73.14KB   51 downloads

Edited by PingPong, 12 March 2014 - 08:16 AM.

[Jiten_process]

Thanks for your detailed reply. Appreciate your time. I was waiting for other experts opinion on this, but I guess I have to conclude now. 

 

I do agree with your point. however, this is not the first time, I have received input compositions component wise previously but in those cases we had been given detail psudo component properties. See attached a typical psudo component properties given for different project. in this case when i model hysys for the compositions given and generated hypo based on psudo coponent properies, It worked well !!!... It is to be noted that particular case (for which the attachment is), c7 to c48 was not mentioned only as "alkanes". So I assume they were meant isomers/aromatics which is inline to what you have explained above. 

 

I was wondering whether the problem is because we are given only molecular weight and density as properties to generate hypo components for C30 to C36+. Had we been given other stanard properties as boiling points and critical temp/pressure (as given in attached case), we would be able to characterise subject crude stream in hysys. 

 

Anyways, I am trying and check with client if they can provide us a detail assay data. 

 

 

what i want to know additionally, can i characterize heavy oil is aspen hysys or i need "upstream" utility of hysys. I do not know bout upstream aspen hysys and hence asking this question. 

Attached Files

•  crude oil data 2.JPG   89.61KB   33 downloads

Edited by Jiten_process, 16 March 2014 - 01:21 AM.

[PingPong]

what i want to know additionally, can i characterize heavy oil is aspen hysys or i need "upstream" utility of hysys.

I use PRO/II, not Hysys, but I am sure that you can enter data as long as you have fraction boiling points, or boiling point curve (TBP or ASTM).

 

I was wondering whether the problem is because we are given only molecular weight and density as properties to generate hypo components for C30 to C36+. Had we been given other stanard properties as boiling points and critical temp/pressure (as given in attached case), we would be able to characterise subject crude stream in hysys.

I think so too, boiling point and specific gravity is all that is normally needed, but without (reliable) boiling points you can't do anything.

 

Anyways, I am trying and check with client if they can provide us a detail assay data.

And then they probably will answer that they gave you a detailed analysis.....

 

Problem is that we don't know how to interpret that analysis, because we don't really know what they mean with C24 or C35, or ......, so we don't know what boiling point to assign to each fraction.

Therefor you best ask them what ASTM (or other) method they used to generate the analysis.

[Jiten_process]

Thanks ping pong for your time and reply. 

[Mattd]

Hi Jitin,

 

I am currently doing the exact same thing as you are attempting to do.

 

I think the underlining conclusion from this, which has previously been stated is that you cannot characterise the crude with the information you have been given. As you have said, previously you have been given properties for the Hypo groups such as NBP, MW, density, viscosity, Critical Temperature, Critical Pressure, Accentric Factor, Watson K, etc.. If you are given these properties then you can completely characterise the crude from a composition as their properties are what a TBP distillation analysis in Oil Manager would provide to you.

 

However, in the case you have now you are not provided with a composition that was generated from a TBP but instead have been given the PVT. As PingPong stated the Carbon Groups above C5 are not just the alkane but aromatics, paraffin’s, and naphthenes. Therefore, simply putting these into components C6-C30 will not provide you with an accurate crude.

 

Normally what you will do is compare the Crude characterised by a TBP with the information you have been given to see if the hypo groups give a good representation. For example checking composition % of the generated hypo group (using the above properties) with the similar boiling point component in the PVT.

 

Hope this helps.

 

Matt