Chemical & Physical Properties of Liquids 1

[Cee555]

I was wondering if anyone knew of a site where the chemical and physical properties of liquids are readily available, I realise that this is probably a massive subject but any help would be appreciated. I work in an industry where quite often I need to know the viscosity, S.G., density...etc.
Perhaps someone knows of a text?

CH

 

[25362]

Try CRC: The Handbook of Chemistry and Physics
or Lange's Handbook of Chemistry

 

[zerok]

Hi Cee555,

try the NIST web book at

http://webbook.nist.gov/chemistry/fluid/

for a an excellent text book see

" The Properties of Gases and Liquids" by Poling et al.

see also

http://www.knovel.com.

Hope this helps

zerok

 

[katmar]

There is a good collection of links to physical properties available on the internet at

http://www.cheresources.com/physinternetzz.shtml

 

[DickRussell]

You should check out the AIChE/DIPPR Project 801 Database (Evaluated Process Design Data). This organization began in 1980, under the umbrella of AIChE. Its first project (801), the data compilation, is ongoing, supported by a great many companies from both the operating and engineering side. Its purpose has been to create the world standard of critically evaluated design data for pure compounds. Part of the original problem was that various sources of data often contained incorrect values or data that were internally inconsistent, and some publications simply replicated data from others without verifying the quality.

Currently, the database covers about 2000 compounds, about 1800 of which are available in the public versions of the database.

Details of the project can be found at

http://www.aiche.org/dippr/projects/801.htm

There are several distributors of the database, with access software in various forms. The present contractor for the project is BYU (

http://dippr.byu.edu).

 

[Gearhead]

Hi
TK Solver has the NIST properties built in to the math modeling tool

http://www.uts.com

 

[jsummerfield]

Neat site for water:

http://www.lsbu.ac.uk/water/

John

 

[elguero]

Perry's Chemical Engineers Handbook has data on many substances, even liquified gases, freons, combustion by products, and mixtures.

Perry's also has a lot of information on equipment such as heat exchangers, piping, corrosion data and such.

 

[TrevorP]

http://www.knovel.com

is a good resource for this sort of information, for a subscription fee. You can trial it for free to see if it has what you need or not. I've often found it to be very useful.

 

[chemELK]

If you are an AIChE member, you can access some areas of the knovel site for free. If interested go to

http://www.aiche.org/membership/knovel/

 

[jlmiller]

Here is a link that I find usefull.

http://infosys.korea.ac.kr/kdb/index.html

 

[UmeshMathur]

The most convenient publicly available tabulation of physical properties data, in my opinion, is the book by Professor Carl Yaws, "Chemical Properties Handbook", (McGraw-Hill, 1999, 1st edition).

Here, you will find the regressed constants for simple equations describing the main temperature-dependent properties (Cp, vapor pressure, etc.) for over 1300 compounds, and also the invariant properties (critical properties, etc.).

Dick Russell has already mentioned the AICHE DIPPR 801 project databank. However, as far as I am aware, this has not been assembled into a book or - preferably - made available in electronically readable form over the internet. This is a great pity, as the projects were originally supposed to make all the developed information public after some years had elapsed.

While these data have been published over time in a number of discrete journals, the consolidated data base has been sold to commercial parties who offer it at very high prices (at least for individuals) and also bundle it with other software in ways that make it hard to decipher the original equation constants. This capability would help immeasurably just about anyone who does any kind of independent work, i.e., unrelated to commercial process simulation packages where DIPPR data are imbedded.

It would be great if the AICHE leadership recognizes that the original intention of the DIPPR projects cannot be achieved unless a decent and inexpensive compilation is made available, at least to all AICHE members.

 

[ddkm]

Umesh,

What a coincidence. I was just using the same handbook (Carl Yaws) last week to obtain certain properties like viscosity and thermal conductivity. For the most part, I got what I wanted and appeared to be quite useful.

However, I was a little apprehensive after going through some of the worked examples. Particularly, because of the fact that his handbook was more of a "correlation" rather than actual empirical data. In fact, the accuracy is openly alluded to throughout the book, example:

a) "The agreement of correlation and data is quite good" or "Correlation and experimental results are in favorable agreement"

b) (From a worked example in viscosity Pg 479) "The calculated and data values compare favourably (... deviation = 3.87%)"

What do you think? "Deviation =3.87%" is not quite "Quite good" to me. Like I mentioned, for the most part I did get the values in the right relevant range for my application, but based on your experiences, how reliable do you think this handbook is, and what are the caveats? Thanks.

---engineering your life---

 

[DickRussell]

For a great many compounds, very little experimental data is available, and when such data exists, often data from multiple sources are not consistent with each other or even within one data set. One of the goals of the AIChE/DIPPR Data Compilation Project 801 that began in 1980 was to provide a compilation of critically evaluated data for use by engineers. Over the last 25 years, highly refined methods have been developed for screening out bad data, performing consistency checks, and predicting values to fill in gaps by the project contractors (originally PSU, presently BYU) and sponsors. The effort is an ongoing one. Along the way, a fair amount of what previously had been available as "data" in the literature has been labeled "not accepted" within the DIPPR database.

Among the data presented for each compound is a quality code, indicating just how good the data are. All experimentally measured data will deviate to some extent from whatever model is chosen to represent the data. One needs to know just how good the data and the correlation derived from it can be expected to be.

Yes, the DIPPR database and access software package costs more than many would like, particularly individuals. But the effort to produce it has been and remains considerable. You get what you pay for. It was never the intent that the results of the project be available to the public for free, just available.

 

[sshep]

Dear All,

Previously mentioned NIST is great for free basic data of pure components

http://webbook.nist.gov/chemistry/fluid/

For mixtures (VLE, LLE, VLLE) Dechema is the largest available data source (can search for free, but lines of data must be bought for small price)

http://i-systems.dechema.de/detherm

Dechema has worked great for me lately. I was recently able to regress binary VLE data for ammonia-ethanol, ammonia-butanol, and other systems that were impossible to find elsewhere.

best wishes, sshep

 

[ddkm]

Dick,

Thanks for the historical background. It does bring a better perspective to what the Carl Yaws handbook is trying to achieve. I was not conscious of the degree of variation in the empirical data. Like most others here (I guess), we were trained to just pluck the data from the tables made available at the end pages of a book! Of course, I would imagine, something like the steam tables would be already quite established and accurate, but I can see your point about other parameters and other compounds.

---engineering your life---

 

[krossview]

Dear All,

Virtual Material have been developing a new process simulator called VMGsim for a while, which is heavily based on the Carl Yaws handbook. It's an easy to use process simulator and fees are reasonable. I've used the simulator for the past four months and have compared the simulation results of VMGsim against Hysys, BRE Prosim and BRE ProMax. The results are within 10%. VMGsim has some neat features, but lack ease of use for case study, Excel import/export and other advanced features when doing detailed studies.

Krossview

 

[UmeshMathur]

ddkm:

Recognize that many of the compounds in the DIPPR database (or any other database) are not found in large quantities in nature. While they can be produced synthetically and purified at great expense, it can literally take weeks of intense effort just to get a few grams of material in the laboratory that is sufficiently pure for measuring properties. Also, several properties cannot be measured with very small amounts of a sample.

Therefore, it is quite common to find pure component properties that have been estimated using a very wide variety of correlations, such as group contribution methods. For many compounds, even decent critical properties cannot be measured, as the compounds decompose at temperatures way below the critical temperature.

Therefore, as Dick Russell has pointed out, one of DIPPR's objectives is to ensure that the values they "certify" for use are consistent with the best practices and procedures. Unfortunately, in the past, a lot of data has been measured with contaminated samples. That is partly what explains the wide spread in reported values. In some instances, the experimental techniques were questionable or of limited accuracy (true for many very old measurements).

There are a number of highly original contributions in the literature (e.g., by Kevin Joback and others) on estimation of properties. The definitive summary is found in Poling, Prausnitz, and O'Connell: "The Properties of Gases and Liquids" (5th edition, McGraw-Hill).

As far as the question of accuracy goes, the pertinent question in every case is: what are you trying to accomplish? For example, it would be foolhardy to insist on extremely accurate pure component vapor pressure data for mixture vapor-liquid equilibria (VLE) in those cases where you have no measured mixture phase equilibrium data (and so must rely on group contribution methods such as UNIFAC for estimating activity coefficients whose accuracy can be off by as much as 20-30%).

In general, if such a question is really critical, one must be ready to go to the literature and try to compare data from a variety of sources, including DIPPR or Yaws, and get some idea of the range of the values that have been reported. DIPPR members also get access to the original literature citations, so the effort to make such a search is rendered less laborious.

Finally, it must be remembered that going back to the laboratory for new measurements is by no means a thing of the past, although I have seen a disturbing trend in recent years to shy away from lab work. Especially for complex mixture VLE work, a few good measurements can be worth a lot more than tons of correlations.

For these reasons, personally, I do not regard any databank as sacrosanct.