160 J . Chem. In/: Comput. Sci. 1991, 31, 160-167

"Val 1 Otol2" (multiplier values for 10-1 2 atoms) and "dec- mark" (leftmost fragment for multipliers from 13 to 19).

The number of terminal symbols that is allowed from a point of failure and, hence, the numbers of reliable interpretations and possiblc correct fragments are dependent on where in a name the failure occurred. For example, if failure occurred within a multiplying term clause, there would be few symbols allowed. If however the failure occurred a t a point where a substitucnt clause could be present, then there would be many terminal symbols to choose from.

In the semantic-correction process, there are several more sophisticatcd investigations that could be added, a t the expense of more complex code and increased processing time. For example, in the case of invalid locants, consider the name 12-dichloropentane. Here, the following deductions could be made conccrning the faulty halogen substituent:

( I ) The locant value of 12 is too large for the parent aliphatic chain length of five carbon atoms.

(2) The explicit multiplier "di" implies that the user probably intended to give more than one locant. (From our experience, it is quite common for mul- tiplier terms to be omitted, but the presence of un- necessary ones is rare.)

(3) A valid name is created if the 2-digit number "12" is converted by insertion of a comma to create two locants, i.e., "1,2- ...".

Thus, the user could be shown the valid name 1,2-dichloro- pentane and asked if that is what was intended.

With this more computationally intensive type of correction, it is important for the system to give up if no solution has been found after some predetermined length of time, or when a certain level of detail has been reached. Nevertheless, as processing power of PCs increases, the number and complexity of attempts to correct erroneous names can expand.

It will be a definite advantage to automatic correction, i f a semantic check is applied where two or more alternative syntactically correct names have been constructed by the syntax error correction system. For example, consider the name pentane-l ,Mol, in which the following two syntactically correct names are automatically produced by the system de- scribed : pentane-l,3-diol (Replacement of "d" by "di")

pentane-l,3-01 (Deletion of "d")

By applying a semantic check on each of these names, it can be seen that while the first is acceptable, the second has a

locant/multiplier term conflict in the position where the syntax correction has just taken place. Thus, both names can be presented to the user, along with a suitable message explaining why the system believes the first to be the more likely can- didate.

The principal conclusion of this work is that it is possible to correct semiautomatically, Le., with some limited input from the user, many of the errors commonly made in the use of systematic chemical nomenclature. An automatic correction mode can provide one or more names that are syntactically correct, but some user judgment is necessary to determine whether these include the intended name.

Applications of the software we have developed could in- clude industrial/commercial data entry and validation of names where input is by nonchemically trained personnel and for computer-based teaching of chemical nomenclature.

ACKNOWLEDGMENT We are pleased to acknowledge the support provided by the

U.K. Laboratory of the Government Chemist and the helpful comments of Mr. I. Cohen of its Chemical Nomenclature Advisory Service on this work.

REFERENCES AND NOTES ( I ) Cooke-Fox, D. 1.; Kirby, G. H.; Rayner, J. D. Computer Translation

of IUPAC Systematic Organic Chemical Nomenclature. 2. Develop ment of a Formal Grammar. J . Chem. Inf Comput. Sci. 1989, 29, 106-1 12.

(2) Cooke-Fox, D. I.; Kirby, G. H.; Rayner, J. D. Computer Translation of IUPAC Systematic Organic Chemical Nomenclature. 3. Syntax Analysis and Semantic Processing. J . Chem. Inf Compul. Sci. 1989, 29, 112-1 18.

(3) Vander Stouw, G. G. Computer Programs for Editing and Validation of Chemical Names. J . Chem. InJ Comput. Sci. 1975,15, 232-236.

(4) Vander Stouw, G. G.; Naznitsky, I . ; Rush, J. E. Procedures for Con- verting Systematic Names of Organic Compounds into Atom-Bond Connection Tables. J . Chem. Doc. 1967, 7 , 165-169.

( 5 ) Vander Stouw, G. G.; Elliott, P. M.; Isenberg, A. C. Automated Con- version of Chemical Substance Names to Atom-Bond Connection Ta- bles. J . Chem. Doc. 1974, 14, 185-193.

( 6 ) Davidson, L. Retrieval of Misspelled Names in an Airlines Passenger Record System. Commun. ACM 1962, 5 , 169-171.

(7) Xerox Corporation. Interlisp-D Reference Manual, Volume 11: En- vironment. Xerox Corp.: 1985; Chapter 20.

(8) Pollock, J . J.; Zamora, A. Automatic Spelling Correction in Scientific and Scholarly Text. Commun. ACM 1984, 27, 358-368.

(9) Lord, M. R. Name Correction and User Interface Enhancements in a Chemical Nomenclature Translator. PhD Thesis, University of Hull, Hull, England, 1990.

(10) Munnecke, T. Give your Computer an Ear for Names. Byte 1980, M a y , 196-200.

( I I ) Damerau, F. J . A Technique for Computer Detection and Correction of Spelling Errors. Commun. ACM 1964, 7, 171-176.

DETHERM: Thermophysical Property Data for the Optimization of Heat-Transfer Equipment

DAVID F. ILTEN

DECHEMA, Theodor-Heuss Allee 25, D6000 Frankfurt am Main 97, Germany

Received October 11. 1990

A physical-property calculation system coupled with a very comprehensive and extensive literature database is described. Methods and methodologies for generating the thermophysical property data necessary for carrying out detailed and accurate design and check-rating calculations of heat-transfer equipment are discussed.

INTRODUCTION generation and heat-transfer equipment are becoming con- tinually more stringent. The "overall" methodologies and algorithms with LMTD values and correction factors tradi- tionally used for the mathematical modeling of fired heaters,

With the ever-increasing sensitivity of economic and pol- lution issues, the requirements for the performance of heat-

0095-2338/91/1631-0160$02.50/0 0 1991 American Chemical Society

DETH ER M J . Chem. If. Comput. Sci., Vol. 31, No. 1, 1991 161

* * * * * * * i n t * * * * * * f n * * * * * * * * . f * * * * t * * * . . * * * * * * * ~ * ~ ~ ~ * ~ * ~ x ~ ~ a * n ~ e ~ ~ ~ ~ ~ ~ a * ~ ~

THREE PHASE LLE FLASH

STREAM 4 TEMPERATURE: 25.00 CEL 298.15 K 77.00 DEG F PRESSURE: 1.0000 BAR 1.0197 KP/CM2 14.504 PSIA CODE NAME MOLE FRACT. MOL.AMOUNT WEIGHT WEIGHT AM.

28 HEPTANE 0.271790 81.537 0.555539 8170.46 120 BENZENE 0.042730 12.819 0.068086 1001.35 10 WATER 0.420594 126.178 0.154559 2273.14

300 ACETONITRILE 0.264887 79.466 0.221816 3262.31 SUM 1.000000 300.000 1.000000 14707.25

* * * * * * * * * * * * * * * * * * * * * f t * * * * * * t f * * * * * * * * a * ~ * * * * * * * * * * * * ~ * n a * ~ * n ~ ~ n ~ x x * ~ ~ * ~ ~ * * a

KMOL FRACTION KG

AVERAGE MOLAR MASS: 49.024 G/MOL MODEL USED FOR LIQUID PHASE NON-IDEALITY: UNIFAC

RESULT OF 3-PHASE CALCULATION:

PHASE 1: 31.324 MOLE % PHASE 2: 32.618 MOLE % PHASE 3: 36.057 MOLE % * THE SYSTEM IS STABLE *

*r*****~~**n*i*r**rrrr********a******aa************************************ THERMOPHYSICAL PROPERTIES OF LIQUID PHASE ? 1 OF STREAM 4

CODE NAME MOLE FRACT. MOL AMOUNT WEIGHT WEIGHT AM.

28 EEPTANE 0.000431 0.040 0.001842 4.06 120 BENZENE 0.000937 0.088 0.003123 6.88 10 WATER 0.767645 72.138 0.590282 1299.59

300 ACETONITRILE 0.230988 21.707 0.404752 891.12 SUM 1 .oooooo 93.973 1.000000 2201.63

* * * * * . * t * * * * a * * m * t f * * * . * * * * * * * * . * * * * * * t . ~ x ~ ~ ~ ~ ~ ~ * * * x * ~ a ~ ~ ~ * * x * * * * x * ~ ~ * * * a ~ x

KMOL FRACTION KG

AVERAGE MOLAR MASS 23.43 G/MOL

THERMOPHYSICAL PROPERTIES OF LIQUID PHASE il 2 OF STREAM 4

CODE NAME HOLE FRACT. HOL AMOUNT WEIGHT WEIGHT AM.

28 HEPTANE 0.828120 81.036 0.878389 8120.23 120 BENZENE 0.121340 11.874 0.100332 927.52 10 WATER 0.002798 0.274 0.000534 4.93

300 ACETONITRILE 0.047741 4.672 0.020746 191.79 SUM 1.000000 97.855 1.000000 9244.46

* ~ * * n * * * * * * x f i f i * * * * * * * * * * * * t * t . * * * * * * * . ~ * ~ * ~ ~ a ~ ~ ~ ~ ~ ~ x n a ~ ~ * ~ ~ n * * n x x ~ a x x ~ a ~ ~

~ * * ~ ~ * * * a ~ * * * ~ x n ~ * ~ * a * * * * * t t * * * L * * t f * * ~ ~ * ~ ~ * ~ * * ~ ~ x ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ x ~ x ~ ~ x ~ x ~ x ~ ~ x x ~ ~ x ~ ~ ~

KMOL FRACTION KG

AVERAGE MOLAR MASS 94.47 G/MOL

THERMOPHYSICAL PROPERTIES OF LIQUID PHASE # 3 OF STREAM 4

CODE NAME MOLE FRACT. MOL AMOUNT WEIGHT WEIGHT AM.

28 HEPTANE 0.004259 0.461 0.014156 46.17 120 BENZENE 0.007924 0.857 0.020532 66.96 10 WATER 0.497045 53.766 0.297018 968.62

53.088 0.668293 2179.40 300 ACETONITRILE 0.490771 SUM 1.000000 108.172 1.000000 3261.14

~ a ~ ~ ~ x n ~ * ~ ~ ~ ~ ~ ~ a ~ x ~ ~ ~ ~ ~ ~ ~ ~ ~ x ~ * ~ ~ x ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ x ~ ~ ~ ~ x ~ ~ ~ ~ ~ ~ ~ x x a ~ x ~ ~ ~ ~ x ~ ~ x

* * * * * * * * * * * * * * * a * * * * * * * * * * f * * * * * n * * * * * x * * * x * * * x * * * * * * * * * x * * * * * * * * * * * * * * * *

KMOL FRACTION KG

AVERAGE MOLAR MASS 30.15 G/MOL Figure 1. Three-phase liquid-liquid-liquid equilibrium flash.

reformers, and heat exchangers are in many cases no longer adequate to meet these more rigorous specifications. Either overdesign will occur, which is wasteful of capital and mate- rials, or underdesign will produce an exchanger which will opcratc at less than optimum conditions. This means loss of hcat throughout the life of the unit. An approach which allows a much more satisfactory representation of the physical reality wi th in these types of equipment and of their mode of func- tioning is the “stepwise” method.’.2 In these methodologies the heat-transfer or heat-generation units are divided math- ematically into a finite number of subunits, each one of which is treated individually and locally as a separate entity. The equations describing each local unit are solved individually and collectively with appropriate boundary conditions. This pro- ccdurc is iterated until the desired convergence of the entire

system is achieved. Certainly a more refined mathematical treatment of the simulation of these devices is of little value if the physical property data necessary for stepwise-type calculations do not also have the appropriate accuracy. Most important is that physical property data be available under the local conditions of composition, temperature, and pressure actually attendant in each of the individual local units of the model employed. In general, tables of values covering the complete range of temperature, pressure, and composition of thc thermophysical and transport properties for the vapor and the condensed states of the pure components and complex mixtures utilized are necessary.

These requisite values of density, viscosity, enthalpy, thermal conductivity, heat capacity, surface tension, and critical pressure are generated by the systems DETHERM-SDR,3”

162 J . Chem. If. Comput. Sci., Vol. 31, No. 1, 1991 ILTEN * . * * * * t * * * * * . . t . t * * * * * . t . * * * * * * * * * * * * * * * * * * * * * ~ * . " * ~ * * * " * " * . ~ ~ * ~ * ~ * ~ ~ * ~ * * * " * * * * ~

VAPOUR-LIQUID EQUILIBRIUM IN BINARY SYSTEM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Y-X-DIAGRAM FOR THE SYSTEM: 2-BUTANOL THE SPECIFIED MOLE FRACTIONS ARE RELATED TO: CALCULATION AFT. UNIQUAC

MOLE FR.1 P = 1.01 BAR VAPOR I

I 1.00 -I 0.98 -I 0.96 -I 0.94 -I 0.92 -I 0.90 -I 0.88 -I 0.86 -I 0.84 -I 0.82 -I 0.80 -I

- HEPTANE 2-BUTANOL

I I I

01 - I-

, 1- * I-

* + I- I- * I-

* + I- I-

* I- + I-

*

*

* 0.78 -I 0.76 -I + 0.74 -I 0.72 -I * + 0.70 -I * 0.68 -I + 0.66 -I + 0.64 -I 0.62 -I + 0.60 -I * + 0.58 -I 0.56 -I * + 0.54 -I 0.52 -I 0.50 -I * + + 0.48 -I 0.46 -I + + 0.44 -1 * + + + + 0.42 -I * + + + + 0.40 -I ++O+ 0.38 -I + + + + 0.36 -I + + 0.34 -I + + * 0.32 -I + + * 0.30 -I + 0.28 -I + * 0.26 -I + 0.24 -I * 0.22 -I + * 0.20 -I + * 0.18 -I 0.16 -I * 0.14 -I + 0.12 -I " 0.10 -I * 0.08 -I + 0.06 -I " 0.04 -I * 0.02 -I 0.00 -IO

+

n + + * +

* + +

I- I- I- I- I- I- I- I- I- I- I- I- I- I- I- I- I- I- I- I- I- I- I- I- I- I- I- I- I- I- I- I- I- I- I- I- I- I- I- I-

T IN CEL

99.6 98.2 97.0 96.0 95.0 94.2 93.4 92.7 92.1 91.6 91.1 90.7 90.3 90.0 89.7 89.5 89.3 89.1 88.9 88.8 88.7 88.6 88.5 88.4 88.3 88.3 88.3 88.2 08.2 88.2 88.2 88.2 88.3 88.3 88.3 08.4 88.5 88.5 88.7 88.8 89.0 89.2 89.5 09.9 90.3 91.0 91.7 92.8 94.1 95.9 98.5

Figure 2. Isobaric binary vapor-liquid equilibrium diagram.

-SDC, and -SDS. DETHERM-SDR provides access to a literature database consisting at present of more than 16 600 documents, 85 500 tables, and 845 500 data records for 3700 compounds. Many of these data, i.e., for VLE, GLE, and transport properties, have been published in the DECHEMA Chemistry Data Series (CDS)."

DETHERM-SDC is a calculational program which allows the determination of properties for complex mixtures. Its internal database contains the pure component data for more than 550 chemical compounds. These data are stored in re- cords of fixed length-the data vectors. Each data vector contains 240 basic data properties for each compound. Ex- amples are density at various temperatures, critical constants, boiling temperature, viscosity, etc. These data are taken from the literature. If several literature values are available for a given property, a data analysis is carried out: The data are carefully evaluated and compared by statistical methods.

Data for all properties for a given pure component in the system are always entered into the data vectors. If a desired

property cannot be obtained from the literature, then it is estimated from the molecular structure or from other al- ready-known properties such as the normal boiling point. DETHERM-SDS is very useful in this regard. DETHERM- SDS provides estimates of data based on the boiling point and whatever other empirical values are available. Of course, the more completely the data vectors are filled with measured values for the individual components, the more satisfactory the estimations are likely to be. For VLE and GLE calcula- tions, appropriate binary interaction parameters are needed for the desired method for calculating phase equilibria, such as Wilson, NRTL, or UNIQUAC. These values are taken from such references as Gmehling et al.' Recommended parameters, which are computed from several literature sources, are available for approximately 140 binary pairs. The interaction parameters for 41 0 other binaries are calculated from reliable single data sets. For the equation of state methods including Lee-Kesler-Ploecker, Redlich, Kwong- Soave, and Peng-Robinson, the binary interaction parameters

DETHERM J . Chem. If. Comput. Sci., Vol. 31, No. 1, 1991 163

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . VAPOUR-LIQUID PHASE EQUILIBRIUM IN THE ACETONE - HEXANE SYSTEM T-X-DIAGRAM FOR THE SYSTEM ACETONE - HEXANE CALCULATION AFT. UNIFAC T I N I P = 0.70 BAR I T IN CEL I I CEL

59.0 -I I- 59.0 58.6 -I I- 58.6 58.2 - I I- 58.2 57.8 - I I- 57.8 57.4 -10 I- 57.4 57.0 -I I- 57.0 56.6 -I t I- 56.6 56.2 - I t I - 56.2 55.8 - I t I- 55.8 55.4 -I I- 55.4 55.0 -I t , I - 55.0 54.6 -1 t I- 54.6 54.2 -I t I- 54.2 53.8 -I * I - 53.8

I I

53.4 -I 53.0 - I 52.6 -I 52.2 -I 51.8 -I 51.4 - I 51.0 -I 50.6 -I 50.2 -I 49.8 -I 49.4 -I 49.0 -I 48.6 -I 40.2 - I 47.8 -I 47.4 -1 47.0 -I 46.6 -I 46.2 -I 45.8 -I 45.4 -I 45.0 -I 41. 6 - I 44.2 -I 43.8 -I 43.4 -I 43.0 -I 42.6 -I 42.2 -I 41.8 -I 41.4 -I 41.0 - I 40.6 -I 40.2 -I

39.4 -I 39.0 - I

I

39.8 -I

t I- 53.4 + I- 53.0

I- 52.6 t I - 52.2 t I - 51.8

*

*

*

* *

* * *

t t

t t

t t

t t

t

t t

t

t

+

I - I - I - I- I - I - I - I - I - I - I- I- I- I - I -

OI- tt I -

t I - t * 1-

t I - t * I -

t I - t * 1-

51.4 51.0 50.6 50.2 49.8 49.4 49.0 48.6 48.2 47.8 47.4 47.0 46.6 46.2 45.8 45.4 45.0 44.6 44.2

43.4 43.0 42.6

43.8

* t t * t * * t t a * t t * ** t * *** t tt ** ****** + + **** * * * * * ~ * * *

I - 42.2 I - 41.8 I - 41.4 I - 41.0 I- 40.6 I- 40.2 I- 39.8 I- 39.4 I- 39.0 I

1 1 1 1 1 1 1 1 1 1 1 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 , O

Figure 3. Isobaric binary boiling temperature diagram,

are taken from Knapp et ala8 The recommended parameters for 130 binaries consisting of low-boiling substances are available.

These data, so generated and appropriately formated, can then be used as input parameters for such simulation packages as SST (Heat Exchangers), FRNC-5 (Fired Heaters), or REFORMER for performing stepwise simulation calculations of high accuracy. These types of calculations are especially important if studies of fouling are undertaken.'2-18 Direct links have been written for such general chemical plant simulation

packages as ASPEN-PLUS.

bilities of DETHERM are given in Figures 1-7. Several examples of data retrieval and calculational capa-

RESULTS

The comparison of heat-exchanger calculations carried out according to the averaging method with those calculated ac- cording to the stepwise method show that there are decided advantages to the stepwise method, assuming that adequate

164 J . Chem. If. Comput. Sci., Vol. 31, No. 1. 1991 ILTEN

* D E T H E R M - S D C * PAGE 1 * DATA CALCULATION SYSTEM VERSION 2 .6 * 04/21/88 .t*.****.a*.*.***.**.t**:*****:a*******************~~:*a**************:*******:**

TABLE AND DIAGRAM OUTPUT

STREAJ 5 t******************t .*~*:*************:a****a***~******~*:*************:~*:**~*

CODE "E VOLE FRAC. MOLE AMObHT WEIGHT WElGTH AM

22 METHANE 0.300000 0 . 3 0 0 0.180613 4 8 1 23 ETHANE 0.300000 0.300 0.338530 9.02 24 PROPANE 0.100000 0.100 0.155482 4.41

M O L FRACTION KG

116 NITROGEN 0.300000 0.300 0.315376 8.10 SUM 1.000000 1.000 1.000000 26.65

* D E T H E R M - S D C P4CE 2 * DATA CALCULATION SYSTEM VERSION 2.6 * 04/21/88 *:*:t~t*****.***.***a***a*****::*******:*****:*:*:*****:******:**~**********~*:

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . TABLE AND DIAGRAM OUTPW

TABLE OF THE VISCOSITIES OF STREAM 5 *****:***:**t*t*.*t************a**::** TEMP. PRES. I STATE I V I S C 0 S I T Y I 1 Y YP4.S I

I I $YERACE U D DEAN I I

CEL BAR I ACGR. I MOLAR AFTER LEE AFTER WILKE AFTER STIEL

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . I I

0.0 1.50 I C I 10.0 1.50 I G I 20.0 1.50 I G I 30.0 1.50 I G I 40.0 1.50 I C I 50.0 1.50 I C I 60.0 1.50 I G I 70.0 1.50 I G I 80.0 1.50 I G I 90.0 1.50 1 c I

100.0 1 .50 I G I 110.0 1.50 I G I 120.0 1.50 I C I 130.0 1.50 I G I 140.0 1 .50 1 G I 150.0 1.50 I G I 160.0 1.50 I G I 170.0 1 . 5 0 I G I 180.0 1 .50 I G I 190.0 1.50 I G I 200.0 1 .50 I G I 210.0 1.50 I C I 220.0 1 .50 I G I 230.0 1.50 I G I 240.0 1.50 I G I 250.0 1.50 I C I

0.01 124 0.01160 0.01196 0.01231 O.Oa266 0.01301 0 , 0 1 3 3 5 0.01369 0.01103 0.01436 0.01468 0.01501 0 , 0 1 5 3 3 0.01565 0.01596 0.01627 0.01658 0.01688 0.01718 0.01748 0.01177 0.01807 0.01836 0.01864 0.01893 0.01921

0.01061 0.01100 0 .01 136 0.01111 0.01206 0.01241 0.01275 0.01309 0.01343 0.01376 0.01409 0.01441 0.01473 0.01505 0.01537 0.01568 0.01599 0.01629 0.01659 0,01689 0.01719 0.01749 0.01778 0.01807 0.01835 0.01864

0.01067 0.01103 0.01139 0.01171 0.01209 0 .01211 0.01278 0.01312 0.01345 0.01378 0.01111 0.01444 0.01176 0.01507 0 .01 539 0.01570 0.01601 0.01631 0.01661 0.01691 0.01721 0.01750 0.01779 0.01808 0.01837 0.01855

Figure 4. DETHERM-SDC viscosity table.

Property

normal boiling point ( K )

critical temperature (K)

DETHERR- L i t e r a t u r e Deviation SDS Value %

352 347 1.2

555 545 1.8

critical pressure (bar) 41.2 44.8 -8.7

standard h e a t of formation, gas, a t 25°C (kJ/Mol) -141.3 -142.4 0.8

density, l i q u i d , at 20°C (gicc) 1.342 1.338 0.3

viscosity, l i q u i d , 2O'C (cP) 0.769 0.824 -7.2

thermal conductivity, l i q u i d , a t 20°C (W/mK) 0.118 0.102 13 .1

Figure 5. I , 1, I -trichloroethane.

Synthesized data compared with literature data for

physical-property data are available for all relevant values of temperature and pressure occurring in the system. Also, a complete, detailed description of the various regions within the heat exchanger or the furnace or the tubes of the reformer can be achieved. This is important for determining eddy currents and studying the special effects which might occur within the piece of equipment.

Stepwise Shell and Tube Exchanger Simulation (SST) can improve the efficiency of heat transfer and monitor fouling in a plant and consequently reduce fuel consumption and flue gas emission from furnaces and boilers. Since FRNC-5 can improve the efficiency of fired heaters and boilers, the resulting optimized designs will release fewer flue gas tons with pollu- tants to the atmosphere. This is especially true for pollutants such as hydrocarbons, CO, and sulfur compounds. This may not apply to NOX because some waste heat recovery with FRNC-5 lowers flue gas temperature and thus increases NO,. However, if NH3 injection is used to lower NO,, the FRNC-5 program can be used to optimize the points of NH, injection and thereby help to reduce NO,.

Those properties which are needed for the calculation are the following: density, viscosity, thermal conductivity, heat capacity, enthalpy, surface tension, and liquid critical pressure.

ND: 0.0-PRP-2.49 COPYRIGHT 06/03/86 DECHEHA E.V.. FRANKFURT/MAIN LA: ENGL DT: 02 ID: CDB AU: FREISER.H.;GLOWACKI.W.L. TI: SOME PHYSICAL PROPERTIES OF ISOQUINOLINE. JA: 3. AH. CHEM. SOC. CO: JACSAT VO: 71 PA: 514 PY: 1949

( 1 ) C9H7N ISOQUINOLINE

NWBER: 119-65-3

COLUMN 1: TEMPERATURE

PURE COMPOUND LIQUID TAG: TSY#ULWNDSTU UNIT: K

COLUHN 2: VISCOSITY

PURE COMPOUND LIQUID

COLUMN 3: DENSITY

PURE COMPOUND LIQUID

TAG: VISIIULLMIDSTU UNIT: PA.S

TAG: DNSNULLMIDSTU UNIT: KG/M3

1: TSY 2: VIS 3: DNS

VALUE

VALUE

VALUE

303.15 0.0032528 313.15 0.0026034 323.15 0.0021323 333.15 0.0017872 343.15 0.001 5269 353.15 0.0013223 373.15 0.0010230 398.15 0.00077870 423.15 0,00062170 448.15 0.00050670 473.15 0.00042230

*** END OF TABLE ***

1091.0 1083.1 1075.2 1067.3 1059.4 1051.4 1035.4 1015.5 994.98 974.21 953.00

Figure 6. Typical document with data on a coal chemical (extract).

Viscosity and thermal conductivity have recently been treated in detail by Stephan and Heckenberger." Especially corre- lalions giving a functional representation of the variation of these transport properties with temperature, pressure, and composition as well as diagrams of the results are presented. Only three methods can effectively describe viscosity over the entire fluid range: 1, the method of Chung; 2, the method of Ely and Hanky; 3, the residual viscosity method. Some il- lustrative examples of the results of Stephan and Heckenberger dcmonstrating the dependencies of these properties on the variables temperature, pressure, and composition are given.

The data for the liquid viscosity of water-ethanol mixtures demonstrate how greatly the viscosity can vary with temper- ature and composition. If the 80% composition line for water is considered, the viscosity varies from roughly 6500 to 632 pPa-s as the temperature is increased from 275 to 350 K. Thus, errors in the position of measuring the temperature can result in large errors in determination of the apparent local temperature. This in turn would lead to very great errors in the viscosity, which brings with it extreme errors in the heat-transfer coefficient of the apparatus. Obviously this will influence the design of the heat exchanger. Assuming an error of 5 K is made in determining the temperature a t 275 K (280 instead of 275 K is measured), then the viscosity of 5039.1 instead of 6481.0 pPa-s, an error of 1447.9 or 22%. For 90% water the percentage error is just as great: 3971.5 instead of 5 1 15.5 pPa-s, and an error of 1 144 pPa-s or 22% as well (cf. p 3 18 in ref 4 and Figure 8). The thermal conductivity also varics greatly. At 275 K it assumes a value of 392.4 mW/m.K for a concentration of 0.8 and 460.0 mW/m.K for a mole fraction of 0.9 in water a t a pressure of 0.1 MPa (cf. p 199 in ref 4 and Figure 9).

This illustrates how essential accurate temperature and pressure values are to the stepwise method. In their treatment of the viscosity of liquids and gases Reid et a1.12 recommend mcthods for correlating viscosities with temperature, estimating viscosities when no experimental data are available, estimating the effects of pressure on viscosity, and estimating the viscosity of mixture^.^ These have partially been incorporated into

DETH ER M J . Chem. If. Comput. Sci., Vol. 31, No. 1, 1991 165

D A T A R A N K : CIIFMSAFF

A U : N a h r r t , K . ; S c h n e n , C . I A : r l r u t s c h e r E i c h v e r 1 a g ; B r a u n s r h w P i ~ ( F r i h . ) r s : 2 .E l - l . PY: 1 9 6 3 - I 9 8 9

N D : 8 . 1 - F X I ~ - 1 0 0 1 . R 9

PIIRE COMPONENT

I J N : 1 1 5 5 : L W : R p g u l a t i r i n s a n d law- c n n c r r n i n g t r a n s p o r t a t i o n a n d t r p a t m p n t

FG: 6 0 3 - 0 2 2 - 0 0 - 4

T h e s e c l a s s i f i c a t i o n s w p r c m a d p h y t h e p r o d u c e r o f t h e d a t a h n s e .

_ _ - - - - _ - - - - _ - - - _ - - - _ l - - - - - l - - - - - - - - - - l - - - - - - - - - - l - - - - - - - - - - - - - - - - - - - - - - - - ! c l a s s ! s u b s . r i s k s ! p a c k . g r o i i p ! n u m b P r / ~ a c k . 111s t I ,

U N O - r e c o m m e n d a t j o n s ! 3 ! ! I ! G C V S p e ( D ) ! 3 . 1 ! ! I ! I MDG-Cod P ! 3 . 1 ! ! I ! GCVE-GGVS ( D ) ! 3 ! ! ! 2 a R D I - A D R ! 3 ! ! ! 2a I C A O / I A T A - r e g u l a t . ! 3 ! ! ! p a s s , p 1 a 11 P : FOR ll I D I I T N

C h e m C - G e f s t o f f v . ( D ) R - p h r a s e s : R 1 2 , 1 9

V . b r e n n b . F I u e s s . ( D ) c l a s s : AI e l e c t r . e q u i p m e n t t e m p ~ r a t u r ~ class ( D I N ) : 4 i g n i t i n n p , r r i i i p ( V D F . ) : 4 e l e c t r . e q u i p m e n t e x p l o s i o n c l a s s : I ~ x p l o s . g i o i t p ( V I l E / D I N ) : I l l 3

! ! ! ! c a r g o p l a n e : 3 0 3

S - p h r a s e s : S 9 , 1 6 . 2 9 , 3 3

: N M : 1 , 1 ' - O X Y R I S E T N A N E : F M : c 4 t i i o n : S D : FORMATION OF P E R O X I D E : C R : 60-29-7

P R O F E R T I E S ! VALUES ! U N I T S - _ _ - - - - - - - _ _ _ _ - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - l - - - - - - - - _ - - - l - - - - - - - -

! MEPt!UUUUUD@TR: M A X I M U M E X P L O S I O N F R E S S U R E , RECOMMENDED ! 10 . O ! R A R

! ! AITOUUUUUF@TR: A U T O I G N I T I O N TEMPERATURE. GERMAN STANDARD, ! 1 7 0. ! C E I. RECOMMENDED ! !

! U E L ~ ~ U U U U U F @ T R : UPPER E X P L O S I O N L I M I T , G E R M A N STANDARD. ! Iloo.!c/n, RECOMMENDED ! !

! ! LELfIUUUUUF@TR: LOWER E X P L O S I O N L I M I T , GERMAN STANDARD, ! 5 0 . ! G / M 3 RECOMMENDED ! !

UEl.#Ul1llUUF(aTR: U P P E R E X P L O S I O N L I M I T , GERMAN STANDARD, ! 3 6 . O ! V O t U M E X RECOMMENDED ! !

! ! LELt jUUUUUF@TR: LOWER E X F L O S I O N L I M I T . GERMAN STANDARD, I 1 . 7 !VOLUME % RECOMMENDED !

! !

I

I

! I

I

FLP!IUUUG#D@TR: FLASH P O I N T , CLOSED C U P , RECOMMENDED ! c -20. ! C E L

Figure 7. Typical document with safety characteristics.

DETHERM. Especially DETHERM-SDS, where thermo- physical properties are estimated for systems where no or only inadequate measured values exist, will employ more accurate mcthods and more efficient algorithms. Future versions of the DETHERM program will be oriented more toward modern interfaces and expert systems to aid the user in se- lecting the methods appropriate to the case at hand, depending upon the temperature, pressure, and composition of the mixture being studied. This is an undertaking which should rightly be shared among the various institutions engaged worldwide in thermochemical research.

The following mixing rules were used in making the cal- culations presented here:

thermal conductivity

viscosity

In Y =

1 (1 - X I b I

A + B(1 - x,)2 x, In y , + ( 1 - xI) In y 2 + In

- (C22)

with x, = mole fraction of water, y , = viscosity of pure water, and y 2 = viscosity of pure alcohol.

I n a database such as DETHERM it can be assumed that there will be an improvement of data quality as a function of time. That is, the reliability of the data increases-let us assume exponentially-as a function of the time that has elapsed since they were entered into the database. A time constant can be defined, y, analogous to a relaxation time which depends (roughly) upon the number of users and the intensity of the use of the data.

The more intensively the data are used, the sooner and the more readily inaccuracies and inconsistencies can be discovered and corrected, whether these stem from the original literature or result from transfer of the data from one storage medium

166 J . Chem. In$ Comput. Sci., Vol. 31, No. 1, 1991

r

Figure 8. Liquid viscosity of water-ethanol mixtures at 0.1 WPa (linear axes) (cf. Stephan and Heckenberger").

p = O , 1 Ml'u, v q w u l i OII ( : 2 1 7Q0

65U

6QQ

550

5no

Y5Q

1100

350

300

250

2LlO

I50

inn 1 , . -- . a . I , 2 . 3 , Y 0 5 $ 6 . l , e , 9 1.0

Mole Fraction Water - Ethanol

Figure 9. Thermal conductivity of water-ethanol mixtures.

I LTEN

Temp. K n 200.00 v zeo.00

A 27b.00

X 300.00 .t 310.00 4 320.00 0 330.00 @ 340.00 X 350.00

to another. This is, then, one of the specific advantages of a database that is used worldwide-the greater the number of users and the greater the variety of problems that are treated,

the sooner inaccuracies will be discovered and the more reliable the database will become. This is also one of the great ad- vantages of the worldwide cooperation among those producing

DETHERM J . Chem. If. Comput. Sci., Vo1.

6500

6000

5500

5000

US00

YO00

4 3500

\ 3000

2500

2000

I500

1000

500

0

cir

,,,,.,'. - .--.. _, _ _ . . - -- -. ._..-. .-e . ---

. a . I , 2 . 3 , Y , 5 . 6 , 7 . 8 ,9 1.0

Mole Fraction Water - Ethanol

Figure 10. Viscosity of water-ethanol mixtures.

and distributing databases. The greater the use and the critical analysis of the database, certainly the more confidence one can have in using the values contained therein.

SUMMARY AND CONCLUSIONS

The more sensitive the mathematical simulation methods, the greater is the required accuracy of the physical-property data. Because the stepwise method yields results for the various regions of the furnaces and heat exchangers where temperature, pressure, and consequently the quality (percent of fluid in the vapor state) vary, it is necessary to reliably know the physical property and especially the transport property and phase equilibrium (VLE) data to high accuracy as a function of temperature and pressure and composition. This applies to the vapor and liquid states as well. The density certainly varies as a function of the temperature and the pressure. As indicated in the figures shown, the viscosity and the thermal conductivities are strong functions of the density. Consequently these properties are also strongly dependent upon temperature and pressure. The stepwise method is inherently more accurate and capable of yielding a more detailed description of the geometry of the heat-transfer equipment and of the internal flow conditions and regimes we11.6-11

It is of little avail, however, if adequate physical-property data are not available. It is the purpose of DETHERM to provide these thermophysical property data under the con- ditions required-whether from the literature, through cor- relations, or through estimation techniques such as incremental methods. Then, with the dual factors of increased mathe- matical simulation capability and adequate knowledge of the requisite thermophysical properties, the total simulation of the heat-transfer system can be carried out to greater accuracy. This allows a significantly more reliable optimization of the equipment so that the desired improvements in economy and environmental protection can be achieved. Thus, the more accurate stepwise simulation method brings advantages for

31. No. 1. 1991 167

I'ernp K A 27600 I 1 2uuoo 0 RRL00 x 20000 I 205.00 0 .IOU (If1 0 :Ill5 uo CD :I I O (10 x 31800 M 320 00 . 325 00

I 33000

\ ? 4 O D O A .I45 00 I I :I50 (10

/ :im ao

the plant operator and for the environment and is a logical starting point for the design of new equipment and the im- provement of existing installations.

REFERENCES AND NOTES

( I ) Ilten, David F. Simulating the Operation and Optimizing the Perform- ' ance of Fired Heaters. Proceedings of the 3rd WIT Conference, June 1985; C.37.

(2) Ilten, David F. Studies in the Stepwise Simulation of Heat Transfer via Air Coolers. Proceedings of the Summer Simulation ConJerence (SCSC) Society for Computer Simulation. San Diego, 1986; p 313.

(3) Ilten, David F.; Eckermann, R. Physical Property Data Estimation Methods in DETHERM. Proceedings of the American Institute of Chemical Engineers National Meeting, San Francisco, 1989; 41 K.

(4) Ilten, David F. Estimation of Physical Property Data with DETHERM. Proceedings of the 12th International CODATA Conference. Colum- bus, OH, July 1990.

( 5 ) Ilten, David F. Thermalphysical Property Data for Heat Transfer Equipment 4. Proceedings of the Rostock Conference on Classical Liquids and Solutions. University of Rostock, Rostock, Sept 1990.

(6) Ilten, David F. Proceedings of the Interfluid 1st International Congress on Fluid Handling Systems; DECHEMA: Frankfurt am Main, 1990; p 635.

(7) Gmehling, Jiirgen, Onken, Ulfert; et al. DECHEMA Chemistry Data Series (CDS) , Vol. I , VLE-DATA, DECHEMA: Frankfurt,

(8) Knapp, H.; et al. Vapor Liquid Equilibria of Low Boiling Substances. CDS VI, VLE for Mixtures of Low Boiling Substances; DECHEMA: Frankfurt, 1982.

(9) DETHERM-SDC User Handbook. DECHEMA: Frankfurt, 1989. (IO) DETHERM-SDR Coaldara User Handbook. DECHEMA: Frank-

furt, 1988. ( I I ) Stephan, K.; Heckenberger, T. Chemistry Data Series, Vol. X, Part 1.

Thermal Conductivity and Viscosity Data of Fluid Mixtures. DECHEMA: Frankfurt, 1989.

(12) Reid, R. C.; Prausnitz, J.; Poling, B. The Properties of Gases and Liquids. 4th ed.; McGraw-Hill: New York, 1987; p 388 ff.

(13) Kern, D. C. Process Heat Transfer. McGraw-Hill Co.: Japan, 1965. (14) PFR, Handbook for Users of SST. Los Angeles, 1987. ( I 5) PFR, Handbook for the Users of FRNC-5. Los Angel=, 1986. (16) Humphreys; Glasgow Handbook for the Users of Heatex. London,

1983. (17) Mendonca, A. J . F.; Nieto de Castro, C. A.; Assael, M. J.; Wakeham,

W. A. The Economic Advantages of Accurate Transport Property Data for for Heat Transfer Equipment. Rev. Port. Quim. 1981,23(7), 1-1 I .

(18) Sieder. E. N.; Tate, G . E. Ind. Eng. Chem. 1936, 28, 1429-1436.

1977-1 990.