• TABLE OF CONTENTS
HIDE
 Title Page
 Preface
 Table of Contents
 List of Tables
 List of Illustrations
 Introduction
 Construction of the rotating concentric-tube...
 Accessories to the column
 Testing and evaluation of the rotating...
 Ocimene
 Summary
 Bibliography
 Appendix
 Biographical note
 Acknowledgement
 Copyright














Title: study of the uses of the rotating concentric tube distillation column.
CITATION THUMBNAILS PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00091327/00001
 Material Information
Title: study of the uses of the rotating concentric tube distillation column.
Series Title: study of the uses of the rotating concentric tube distillation column.
Physical Description: Book
Creator: Burris, William Alan,
Publisher: University of Florida
 Record Information
Bibliographic ID: UF00091327
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: alephbibnum - 000546240
oclc - 13207638

Table of Contents
    Title Page
        Page i
    Preface
        Page ii
    Table of Contents
        Page iii
        Page iv
    List of Tables
        Page v
    List of Illustrations
        Page vi
        Page vii
        Page viii
        Page ix
    Introduction
        Page 1
        Page 2
        Page 3
        Page 4
        Page 5
    Construction of the rotating concentric-tube distillation column
        Page 6
        Page 7
        Page 8
        Page 9
        Page 10
        Page 11
        Page 12
        Page 13
        Page 14
        Page 15
        Page 16
        Page 17
        Page 18
        Page 19
        Page 20
        Page 21
        Page 22
        Page 23
        Page 24
        Page 25
    Accessories to the column
        Page 26
        Page 27
        Page 28
        Page 29
        Page 30
        Page 31
        Page 32
        Page 33
        Page 34
        Page 35
        Page 36
        Page 37
        Page 38
        Page 39
        Page 40
        Page 41
        Page 42
        Page 43
        Page 44
        Page 45
        Page 46
        Page 47
        Page 48
        Page 49
        Page 50
        Page 51
        Page 52
        Page 53
        Page 54
        Page 55
        Page 56
        Page 57
        Page 58
        Page 59
        Page 60
        Page 61
        Page 62
        Page 63
        Page 64
        Page 65
        Page 66
        Page 67
        Page 68
        Page 69
    Testing and evaluation of the rotating column
        Page 70
        Page 71
        Page 72
        Page 73
        Page 74
        Page 75
        Page 76
        Page 77
        Page 78
        Page 79
        Page 80
        Page 81
        Page 82
        Page 83
        Page 84
        Page 85
        Page 86
        Page 87
        Page 88
        Page 89
        Page 89a
        Page 90
        Page 91
        Page 92
        Page 93
        Page 94
        Page 95
        Page 96
        Page 97
        Page 98
        Page 99
        Page 100
        Page 101
        Page 102
        Page 103
        Page 104
        Page 105
        Page 106
        Page 107
        Page 108
        Page 109
        Page 110
        Page 111
        Page 112
        Page 113
        Page 114
        Page 115
        Page 116
        Page 117
        Page 118
        Page 119
        Page 120
        Page 121
        Page 122
        Page 123
        Page 124
        Page 125
    Ocimene
        Page 126
        Page 127
        Page 128
        Page 129
        Page 130
        Page 131
        Page 132
        Page 133
        Page 134
        Page 135
        Page 136
        Page 137
        Page 138
        Page 139
        Page 140
        Page 141
        Page 142
        Page 143
        Page 144
        Page 145
        Page 146
        Page 147
        Page 148
        Page 149
        Page 150
        Page 151
        Page 152
        Page 153
        Page 154
        Page 155
        Page 156
        Page 157
        Page 158
        Page 159
        Page 160
    Summary
        Page 161
        Page 162
        Page 163
        Page 164
        Page 165
        Page 166
    Bibliography
        Page 167
        Page 168
    Appendix
        Page 169
        Page 170
        Page 171
        Page 172
    Biographical note
        Page 173
    Acknowledgement
        Page 174
        Page 175
    Copyright
        Copyright
Full Text










A STUDY OF THE USES OF THE ROTATING

CONCENTRIC TUBE DISTILLATION COLUMN






By
WILLIAM ALLAN BURRIS


A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF
THE UNIVERSITY OF FLORIDA
IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE
DEGREE OF DOCTOR OF PHILOSOPHY









UNIVERSITY OF FLORIDA
AUGUST, 1955

















PREFACE


"Now I am come to the arts and I shall begin
from distillation, an invention of later times, a won-

derful thing to be praised beyond the power of men; not

that which the vulgar and unskilled men use, for they do

but corrupt and destroy what is 'ood; but that which is

done by skillful artists.... Lot one that loves learning

and to search nature's secrets, enter upon this; for a

dull fellow will never attain to this art of distilling."(a)


(a)Dlla Porta, MaHia Naturalis, Naples, 1589; quoted in
Egloff and Lowry, Ind. Eng. Chem., 21, 920 (1929).












TABLE OF CONTENTS


Page
PREFACE ......... ..... ..... .., ...... ......... ... ..

LIST OF TABLES....... ........ ............. ... ... .... v

LIST OF ILLUSTRATIONIS ............ .. ....... ... . vi

CHAPTER

I. Introduction...... .................I.... o.... 1

II. Construction of the Rotating Concentric-
Tube Distillation Colu6mn.................. 6

III. Accessories to the Column..o.................... 26

Power System ... ........... ............ .... 26

Heater Control...2.................... ... 28

Thermocouple System......*..... ..*.... 32

Throughput Control......................... 36

Column MNotor As sembly-.................. 5

Tachometer......... ......... ............ 6

Vacuum Systemr................ ..... ...... 58

Distillation Head........................ 61

IV. Testing and Evaluation of the Rotating Cclumn. 70

Test Apparatus and Sampling............... 87

Operation of the Column and Test Procedure. 90

Testinr at Atmospheric Prescure...... *..., 95

Evaluation of Atmospheric Pressure Tests.,. 101

Testing and Evaluation of the Rotating
Colurmn Under Reduced Pressuro..o........... 116


iii









TABLE OF CONTEiTS (Continued)


CHAPTER

V. Ocimene...., .. ............... .,,.,.......... 126

Preliminary Experiments.........,......... 130

OcIziene Apparatus.**..,.................... 137

VI. Summary.* .......*...,.....*.. .... ..... *....** 167

BIBLIOGRAPHY...o.... .............*.... **.... ........ 173
APPEN-DIX. ................ ............................ 175
173
ACKOWLEDGMENTS...................................... 13o 4 0oo .i79

BIOGRAPHICAL NOTE...................................... 180

C1OMIITTEE REPORT....................................... 181













LIST OF TABLES


Table No. Page

I. Data for the Relation Between Composition
and Refractive Index of the Mixture
n-heptane and Dibutyl Sebacate................. 80

II. Holdup of Rotating Column Test Mixture
n-heptano Dibutyl Sebacate...............,****** 85

III. Rotating Concentric-Tube Distillation Column,
Atmospheric Pressure, n-hoptane and Mothyl
Cyclohexane = 1.070, Rectifying Section
125 cm. Total Reflux........................... 114

IV. Test Data for Rotating Concentric-Tube Distil-
lation Column at 200 ran. of HI Condenser
Pressure; Test Mixture, n-heptane and Methyl
Cyclohexano.... ..... ............... .... ...... 123

V. Test Data. Effect of Pressure on Rotating
Concentric-Tube Distillation Column Efficiency.. 125

VI. Optical Density at Characteristic Wave Lengths
of Pure Components of Pyrolysis'Mixture in
.0288 im. cell.................................. 152

VII. Optical Density of Pure Components of Pyrolysis
mixture in 0.0288 cell at Infrared Wave Lengths
Most Useful in Quantitative Determinations...... 154

VIII. Compositions of Mixtures Obtained at Different
Voltages Across the Pyrolysis Spiral............ 155













LIST OF ILLUSTRATIOITS


Plate No. P

I. Rotating Concentric-Tubo Distillation
Column, Upper End..........................

II. Rotating Concentric-Tube Distillation
Column, Lower End......,.................

III. (a) Oil Throw Ring..........................
(b) Top End Cap.............................

IV, (a) Upper End Cap and Rotating Shaft Seal
Assembly..............o..........*......
(b) Upper End Cap and Stationary Seal Plate

V. (a) Top View of Upper End Assembly Showing
Rotating Seal in Oil Chamber............
(b) Bearing Housing mnd Oil Chamber.....,...

VI. (a) Bearing Lock Nut and Oil Retainer.......
(b) Top View of Upper End Assembly,
Showing External Bearing..............

VII. (a) Upper End Assembly......................
(b) Bottom View of End Cap...............

VIII. (a) Bearing Support Block.........,....
(b) Side Arm....,...........................

IX. Bottom End Cap and Attachment Tube..........

X. Upper End of Rotating Column with End Cap
Removed and Rotor Pulled Out...............

XI. Control Panel........,...o......*.......

XII, Power and Heater Circuits..................

XIII. Thermocouple System........................

XIV. Optical System............................

XV. Thermocouple Calibration, Copper-Constantin
Couples in Positions 1-5....................

XVI. Thermocouple Calibration, Temperature vs.
Millivolts Per Degree......................










ILLUSTRATIONS (Continued)


Plate No.


Page


vii


XVII. Thermocouple Calibration Correction........

XVIII. Thyrqtron Temperature Differential Through-
put Control Circuit.........................

XIX. Inlet Side Armi of High Speed Condenser......

XX. Water System...... ..... .... ... .... .........

XXI. (a) Motor Circuit... ............. ...........
(b) Timing Circuit................ .... .......
XXII. Motor Assembly,..............................

XXIII. Vacuum System....... ...... .... .............

XXIV. (a) Automatic Vapor Dividing Head...........
(b) Details of Vapor Dividing Valve.........

XXV. Kettle Amperos vs. Thcrmocouple E.I.F. in
Millivolts ......... .... ............... .....

XXVI. Percent Dibutyl Sebacate Against Refractive
Index................. ..................
XXVII. Holdup Apparatus...........................

XXVIII. Effect of Throughput on Rotating Column
Holdup.......* ...... ... .. ...... ... .......
XXIX. Test Kettle with Vacuum Sampling Device.....

XXX. Control Panel................... ............

XXXI. Mole Fraction of n-heptane in n-heptane and
'Methyl Cyclohexane Mixture Against Refractive
Index at 25Centigrade......................
XXXII. Equilibration Curve for Rotating Concentric.
Tube Distillation Column at 5000 RoP.M. with
a Throughput of 3.4 Molos/Hr. in Test Mix-
ture: n-heptane and Methyl Cyclohoxano.....
XXXIII. Theoretical Plates of the Rotating Concentric-
Tube Distillation Column Against R.P.M. at
Several Values of Throughput. Test Mixture:
n-heptane-methyl cyclohexano, Condenser
Pressure, 760 mm. ig........................


LI.0

45


51

52

57
57
59
62

67
67


74

81

83

86

89

94


96



97




107











ILLUSTRATIONS (Continued)


Plate Ho. Page

XXXIV. Theoretical Plates of the Rotating Concen-
tric-Tube Distillation Column Against
Throughput in Moles Per Hour at Several
Values of R.P.M. Test Mi.;ture: n-heptane
and iethyl Cyclohexane. Condenser Pressure,
760 ram. iH .................................. 108

XXXV. Pressure Drop in Millimoters of Ii in the
Rotating Concentric-Tube Distillation Column
Against R.P.M. for several values of Through-
put .................... ................... 110
'XXXVI. Pressure Drop in Millimeters of Hg in the
Rotating Concentric-Tube Distillation Column
Against Throughput in Moles Per Hour for
Several Values of R.P.M.1................... 112

XXXVII. Theoretical Plates in the Rotating Concen-
tric Tube Distillation Column Against R.P.M.
at Three Values of Throughput. Condenser
Pressure 200 min. Hg. Test Mixture: n-heptane
and M;ethyl Cyclohexane ............* .......*. 120

XXXVIII. Theoretical Plates in the Rotating Concen-
tric-Tube Distillation Column Against
Throughput in Moles Per Hour at Several
Values of R.P.M. Condenser Pressure, 200
mm. of Hg. Test Mixture: h-heptane and
Methyl Cyclohoxane............... ....... 121

XXXIX. Effect of Condenser Pressure on Efficiency
of the Rotating Concentric-Tube Distil-
lation Column.................... p........ 124

XL. Formulaso............ ..... ....... ...... .. 129

XLI. Apparatus for Production of Ocimene by
Vapor Phase Pyrolysis of Alpha Pinene....... 135

XLII. Top View of Ocimone Apparatus.*......*....... 11

XLIII. Water Jacketed Ocineno Condenser and Re-
ceiver... *o. ... *.. .. .......... .,........., 1 2


viii












ILLUSTRATIONS (Continued)


Plato INo.

XLIV. Oc imone Apparat ..........................

XLV. Infrared Spectrum of Alpha Pinene...........

XLVI. Infrarod Spectrum of Alloocimiene D1'ior......

XLVII. Infrared Spectrtu of Alloocirnon from iHct
Tubc Vapor Phase Pyrolysis at 350C........o

XLVIII. Infrared Spectrum of Dipentene...........oo

XLIX. Infrared Spectrum of Alloocimene from Vapor
Phase Pyrolysis Over Hot Spiral..........*o

XLX. Infrared Spectrum of 91O, Ocioenc, 6/ Di-
pentene Nixture .............................

XLXI. Infrared Spectrumn of Synthotic Pyrolysis
Mixture.................... o................


Page

143

145
146


147

148

149


150
















CHAPTER I


INTRODUCTION

The present development of the rotating concen-

tric-tubo distillation column as a more practical device,

came as a ro~elt of difficulties experienced in the sepa-

ration of close boiling, thcrmolabile terpenes, particu-

larly ocimeno.. Existing columns had the disadvantages of

high pressure drop, high holdup, and very low throughput.

The rotating column, as originally described by Rossini,

Willinghami, Sodlak, and Westhaver(1 appeared to greatly

reduce these disadvantages. Since the original work at

the Bureau of Standards, there have been no additional

publications confirming or oxpandicn their findings.

The National Bureau of Standards' column had a

rotor consisting of a closed cylinder 23.0 inches (58.4 cm.)

in length and 2.93 inches (7.111 cm.) in outside diameter.

The rectifying section consisted of the empty annular

space, 0.0113 inches (1.09 mm.) in width, between the rotor

and the outer, stationary tube. The rotation of the inner

tube caused the laminar flow, characteristic of stationary

concentric-tube columns, to chance to turbulent flow.

Since this increased the diffusion coefficient in the











vapor phaso, a greater roctifying efficiency was obtained.

The column showed very slight increase in efficiency

up to 2300 R.P.M., at which speed the chanfre to turbulent

flow apparently was made. Above 2300 R.P.M. there was a

marked beneficial effect of increased R.P.M. on efficiency.

The theoretical treatment predicted that after an initial

sharp increase in efficiency at the onset of turbulence,

greater speed of rotation would not produce a much greater

efficiency. However, experimentally it was found that at

least up to 1-000 R.P.M., the highest speed investigated,

the efficiency was rauch greater than predicted by theory

and that the marked beneficial effect of rotation con-

tinued to increase with increased rotation. This unex-

pected result was explained on the basis of eddy diffusion

in the film of liquid reflux. The experimental work also

showed that a decrease in throughput greatly increased

the platage of the rotating column. Data showed very low

values for the calculated holdup and the experimental

pressure drop. The holdup and pressure drop wero increased

by an increase in throughput. The pressure drop was in-

creased by increasing R,P.M.

At 41000 R.P.H. and a throughput of 2000 to 4000

ml./hr., the National Bureau of Standards column had an

efficiency factor of about ton times that of any previously

reported column. The ef iciency factor is the throughput










divided by the holdup per equivalent theoretical plate

and is e-xressod in units of the number of equivalent

theoretical plates through which the material passes in

unit time.

It was desired to reduce the time of distillation

required for the separation of close boiling terpones by

designing a column with a sufficient number of plates

which would operate successfully at a greater throughput

rate than obtainable with the existing column. In addition,

in some cases the packing in the available columns caused

isomerization and polymersiztion of the materials, due to

the prolonged heating. The higher values of pressure drop

of several of the conventional columns caused the kettles

to be operated at a higher pressure and, therefore, at a

higher temperature, which favored isomerization and polymeri-

zation. The higher values of holdup meant that a large

portion of smaller samples were lost. The rotating con-

centric tube column offers a low pressure drop, low hold-

up and about 20 times the throughput of existing columns at

comparable HETP's. The National Bureau of Standards column

was designed only for manual sample taking at atmospheric

pressure. For the purpose of distilling torpenes the

rotating column was redesigned for automatic collection of

fractions and for vacuum operation. At the same time an

attempt was made to increase the efficiency and the total

platage, by modification of the dimensions.










The purposes of the research described in the

following chapters are:

(1) To design and construct a rotating concen-

tric-tube distillation column capable of vacuum operation.

This is discussed in Chapter II.

(2) To modify original dimensions of the National

Bureau of Standards column in order to improve the efficiency

and the total platage. This is also discussed in Chapter II.

(3) To design and construct the auxiliaries
necessary for automatic fraction collecting at the high

rates of throughput of which the column is capable of

being operated, and for the efficient operation of the

column. This is discussed in Chapter III.

(LI) To test the column to confirm the previously

reported characteristics. This is discussed in Chapter IV.

(5) To test the column to evaluate the effect
of the modifications of dimensions. See Chapter IV.

(6) To extend the results to higher R.P.M.'s,

to determine if the efficiency continues to increase

indefinitely with increasing speed, or whether an optimum

speed exists. This is discussed in Chapter IV.

(7) To determine if test results at atmospheric
pressure apply equally well at reduced pressures. See

Chapter IV.

(8) To demonstrate the value of the rotating

column in the separation of terpones, and particularly











ocimene from the compounds formed in its production.

(9) To design equipment for the efficient prepara-

tion of ocimeno.

(10) To determine certain properties of ocimene.















CHAPTER II


CONSTRUCTION OF TIIE ROTATING COICENITRIC TUBE
DISTILLATION COLUMN


In order to provide for vacuum operation and

automatic take-off, the original Rossini column had to

be extensively redesigned. The original Rossini column 9)

had a rotor consisting of a closed cylinder 23 inches

long and three inches in diameter. The rotor turned in

a stationary tube with 1.09 millimeter annular clearance.

To increase the peripheral velocity of the rotor in the

present model, the diameter was increased to four inches.

The annular clearance was cut to 0.032 inch (0.8 mm.)

since Rossini's article indicated that efficiency might

be increased in this manner. The total platage was in-

creased by increasing the length of the rotor to 50

inches (125 cm.).

The rotating column is supported by two radial
(a)
thrust bearings (See Pg. 16, P1. II, Pt. 5) press

fitted in a bearing support block at the bottom of the

column. The support block (Pg. 16, P1. II, Pt. 8; Pg. 23,


(a)g. is page number of plate, P1. is plate number, Pt.
is part number.









7
PI. VIII, a) is designed to cause so1'.e of the liquid reflux

to run through holes in the support block. The block it-

self is threaded into the end of the outer stationary tube.

The upper bearing support block (PC. 14, P1. I,

Pt. 13) has its shoulder above the single self-aligning
bearing (Pg. 14, Pl. I, Pt. 14). Since the shaft (Pg. 14,

PI. I, Pt. 1) has a shoulder against the bottom of the

bearing, end play is prevented. Both the upper and lover

bearing support blocks have an Allyn sot screw to anchor

the threads. The upper bearing support block is designed

to allow liquid reflux to lubricate the upper bearing in

the same manner as the lower bearing. The bearings and

the bearing support blocks are accessible by unscrewing

the end caps, (Pg. 11, P1. I, Pt. 1C; Pg. 16, P1. II,

Pt. 4; Pg. 22, P1. VII, b). Spanner wrench holes are
provided in the end caps for easy removal. A special

spanner type wrench, with two pins perpendicular to the

plane of the wrench, is used to remove the bearing sup-

port blocks. Allyn set screws prevent the bearing

support blocks from unthreading while in operation. In

addition, the top shaft is supported by an external, oil

lubricated, self-aligning ball bearing, (Pg. 14, P1. I,

Pt. 3; Pg. 21, P1. VI, b). These self-aligning bearings

have good high speed characteristics and avoid the

aligning difficulties experienced previously with rigid

bearings. The decision to add the external bearing, to








8
change from rigid to self-aligning bearings, and to increase

the shaft size from five-eighths inch to approximately one

inch was made after the crystallization and complete

breaking of the original shaft. These modifications have

so far prevented a recurrence of this incident.

Balancing of the stainless steel rotor (Pg. 14,

P1. I, Pt. 16; Pg. 25, PI. X) to reduce vibration proved

to be a major difficulty of fabrication. After the rotor

had its end plates and shafts attached and had been machined

down to the correct diameter, it was statically balanced.

However, because of the unevenness of the stainless steel

tube from which it was fabricated, it was still far from

being in dynamic balance. The dynamic balance was im-

proved by having the column balanced by the General Elec-

tric Shop in Atlanta, Georgia. Dynamic balancing might be

avoided by fabricating the rotor from solid aluminum or

magnesium or by machining the inside of a hollow rotor.

Elimination of stainless steel would also greatly improve

fabrication from the machinist's viewpoint. If these

metals were used, however, a new problem of how to attach

the shafts would be introduced. A better idea of the

magnitude of the problem of balancing a rotor of this

size is given by the fact that even after dynamically

balancing, bolts in the preliminary test stand crystal-

lized and broke. However, the second and larger diameter










shaft and improved bearing system, plus shock mounting,

greatly reduced the vibration and its effects. This

shock mounting consisted of bolting the flange (Pg. 22 ,

P1. VII, b) at the top of the column to the top of a

reinforcing plate attached to the large heavy cabinet

used as the column support, with 1/8" rubber between the

flange and the plate. In addition, the same thickness of

rubber was inserted between the reinforcing plate and the

top of the cabinet. The bottom of the column was held by

three springs (Pg. 24, P1. IX) bolted to a flange which

was welded to the bottom end cap. The other ends of the

springs were fastened to rods bolted to a shelf within

the cabinet.

To prevent any possible damage to the glass equip-

ment attached to the column, flexible metal couplings

(Pg. 16, P1. II, Pt. 10; Pg. 23, P1. VIII, b; Pg. 24,
P1. IX) were introduced into the side arm tube at the

top (Pg. 24, P1. IX) and into the bottom attachment tube

(Pg. 16, P1. II, Pt. 10; Pg. 24, P1. IX). These flexible

couplings plus the use of metal semi-ball joints con-

necting with glass semi-ball joints, provided flexible

and vibration-free metal to glass connections. The top

and bottom end caps are threaded onto the column. They

are vacuum sealed by permatex gasket compound on the

threads, and by putting the caps down right against the










gaskets which are supported by rings (Pg. 16, P1. II, Pt. 2)

shrunk on the stationary stainless steel outer tube (Pg. 14,

P1. I, Pt. 15; Pg. 16, P1. II, Pt. 7). The side arm,

which carries the vapor and liquid between the top of the

column and the head, is attached by threads and vacuum

sealed by permatex compound on the threads. Permatex is

a commercial product resistant to hydrocarbons. The main

sources of vacuum leads were; the threaded connection

between the side arm and the head, between the column and

,nd caps, and the pipe fittings on the bottom cap. Perma-

tex eliminated the side arm and end cap leaks and silver

soldering eliminated the pipe fitting leaks. An unusual

type of vacuum leak was presented by a pin hole in the

hollow rotor. This pin hole leaked the air from the

rotor into the vacuum system fast enough so that it was

impossible to pull a good vacuum, but slowly enough so

that several days were necessary to completely evacuate

the column.

By far the most difficult task in the construction

of the column was the problem of vacuum sealing a shaft

rotating at speeds up to 6000 R.P.M, The idea of using

an enclosed or induction motor to avoid the sealing

problem was dropped because of the probability of arcing

in the vacuum and the problem of cooling a motor insulated

by the vacuum. Three different shaft sealing systems were










tried before arriving at a satisfactory solution. First

tried was a packing gland with graphite-covered asbestos

cord as packing. This failed because the shaft was a few

thousandths of an inch out of line. An external bearing

would have improved this seal. The second attempt em-

ployed a sylphon seal lubricated with a mixture of cup

grease and molybdenum sulfide. This maintained a satis-

factory vacuum, but slowly leaked grease and molybdenum

sulfide into the column. This material was non-volatile

and did not interfere with the distillation, but the

contamination did make analysis of test mixtures in the

pot difficult. A shutdown period during the repair of

the broken shaft was utilized to add the external bearing

and the improved oil lubricated shaft seal. This new

shaft seal arrangement had the advantages of using centri-

fugal force to oppose the leaking of oil and permitting

the immersion of the seal faces in oil for cooling and

lubrication. As a safety device, an oil throw ring and

an oil drainage system were added. The external bearing

housing serves the extra purpose of containing the lubri-

cating oil for both the seal and the external bearing.

This housing is aligned with the shaft by the use of

tapered pins. Jacking screws are provided for removal

of housing and bearing. The column temperature is regu-

lated by three independently controlled heaters. These

heaters consist of number 20 chromel-A resistance wire












lathe wound on an asbestos paper covered sheet metal tube.

The tube was made by soldering a rolled piece of sheet

metal which was wrapped around t1h column and held by

metal straps during soldering. The ends of the heater

wires are held by teininals threaded into pieces of

micata insulation. The micata blocks are attached with

screws to the flat tops of pieces of steel cut on the

bottom to match the sheet metal tube radius. The steel

pieces are soldered to the sheet metal. The one-fourth

inch wide air space between the heater and the outer

tube of the column is divided into three sections by two

brass rings. These sections correspond to the three

heater sections. The sheet metal tube is attached to

these rings by screws. Three one-inch circular holes

were made in the sheet metal tube to allow placing of

thermocouples between the heaters and the outer tube of

the column. Eight flat-headed screws were soldered

on the sheet metal tube around each hole. They were ar-

ranged four on the right side and four on the left side

of the hole in a square pattern. The thermocouple wires

were held in place by tying them to these screws with

thin wire. The asbestos paper was glued to the sheet

metal tube with sodium silicate. The side arm heater

consists of number 22 nichrome wire wound on three-

fourths inch thick corrugated asbestos steam pipe lagging.









13

Both the column heaters and the side arm heater, as well

as the unheated bottom attachment tube, are covered with

glass wool and wrapped with asbestos cloth tape.








PLATE I











414





7
-- D



















F







I i
I^ 0






Scale 1/2 1

Rotating Concentric Tube Distillation Column
Upper End








PLATE I


Sha Top 3/4" is 1/2" diameter. Shaft is
0.943" at bearings. Between bearings 31/32"
Welded to rotor end plate (P!. 25, PI. X)
O Bearing lock nut and oil retainer (Pg. 21, P1. VI a)

O Ball bearing, self aligning 0.5906" x _.O072".
I.D. 0.9843. S.K.F. bearinC No. 7205BDT (PC.21,P1.VI b)
Rotatin .face of rotating shaft seal assembly
Parr lNo. 6357 for 31/32" shaft (Pg. 19, PI. IV a)
Oil filling tuba. 1/8" pipe. Connected to oil
reservoir by plastic tubing. (PC. 20, Pl. V b)
SSpring, rotating shaft seal. (Pg. 19, Pl. IV a)

7 -External bearing housing and oil chamber
(Pg. 20, Pl. V b)

Oil throw ring (PC. 13, Pl. III a)
SDrain hole, oil chain well. Threaded for 1/4"
pipe. (Pg. 18, Plate III b)
SCap, upper end. Treadoe 1 1/2". (Pg. 18, Pl.
III b and Pg. 22, Pl. VJI b)
O Stationary seal plate, rotating shaft seal as-
sembly. Parr No. 6357 (Pg. 19, P1. IV b)
SOutlet to condenser, threaded for 1" pine
(Pg. 1, P1. III b and Pg. 22, P1. VII b)
Upper bearing support block. 1.145" x 4 1/16.
I.D. 2.0472~. (g. ( 23, Pl. VIII a)
Self aligning ball bearing, same as (3)

O Outer concentric tubc, stainless steel
53 1/2" x 4 1/2". I.D. 11.025
SRotor, stainless steel, 50" ;: 3.962. 3 1/2" I.D.
1/2" plates welded on each end.







PLAlE II


--- I-I- s _2'-
-I- 181-- '1


Rotating Concentric Tube Distillation Column
Lower End


2 1 24







PLATE II


LEGEND

O Brass ring, split, 1/2" x 5", I.D. 4 1/2"

O Steel ring, 3/8" x 5", I.D. approx. 1 1/2" (Pg.25,P1.X)

O Bearing press fitted

O Cap, lower end threaded 1 1/2" (Pg.24 P1. IX)
Bearings duplex radial thrust, 1.181" x 2.01:72",
I.D. 0.98~3". S.K.F. bearing number 1205
O Shoot metal cylinder, 53 1/2" x 5".
SOuter concentric tube. Sane as 15 in Plate I
(Pg. 1I, P1. X)
Lower bearing support block 1.392" x 4 1/16",
I.D. 2.01!72 (Pg. 23, P1. VIII)
O Flexible metal coupling (PC. 24, Pl. IX)

Flexible metal coupling (Pg. 24, P1. IX)

O Flexible metal coupling (Pg. 24, P1. IX)

O Sleeve, 1" pipe (Pg. 2L!, P1. IX)

OQ iotal Somiball joint 50/30 (Pg. 24 P1. IX)










PLA2 III, aL


Oil ThJrow ling

'IA:' III, b


Top Lnd Cap










PL/." .. V,


Uppor ',.nd Car p r-id x otcvti.r Zj.t ;"ocl Asa..nbly

PIP-'.. IV b


Upper 2;nd Cap and >>tatio:n..-7 oa1l late











PLATW V., a


S


Top View of Upper End Assembly Showing Rotating Seal in
Oil Chamber

PLATE V, b


Bearing iiousing and Cil Chamber











?'7 1 VI, a


r lrx


Boori.1r4 LocI: jiut and Oil IRotc.i:r


2L2,.A VI, b


Top View ofi Upper -;nd j.ssor:ilys, Shlcin ixtoernal DerlrLngC










PLA T7 VII, a


Uppor 'nId Assembly

PI -2- VII, b


Bottom View of c-d Cap










PLAL- VIII, a


Dearirn Jupport J3lock


PLAT.- VIII, b


Slide tAn












PLL.'W IX


Jottom End Cap and Attachment Tube













PLATE X


Upper and of Rotating Column with End Cap Removed and
Rotor Pulled Out
















CHAPTER III


ACCESSORIES TO THE COLU1,i


After construction of the column, the next most

important problem was to design the control system and

construct the auxiliaries necessary for efficient opera-

tion. Without the proper instrumentation to guide the

control, it would be difficult to obtain the best per-

formance. To centralize and organize the controls, a

desk type control panel was built of angle iron and

masonite. This control panel is shown on Page 27,

Plate XI, and Page 37, Plate XIV. The interior of the

panel is accessible through a hinged side door, or

through a sliding back panel.


Power System

Since the column is operated primarily elec-

trically, the power system will be discussed first.

A diagram of the power circuit is shown on Page 30,

Plate XII. The power comes to the panel in a conduit

containing three wires. Two wires are load lines and the

third a common. Between either of the load lines and

the common,115 volt potential is available. Between

26








PLA'E XI


I I' '


Control Panel












the two load lines a potential of 208 volts is available

but not used. These lines are run first through a main

switch, which is accessible through the side door of the

panel. From the main switch the power goes through a

main magnetic switch, which is operated by push buttons

on the front of the panel. A load line and a cormon line

from the magnetic switch are attached to a 2 K.V.A. Sola

voltage regulator, which is connected to the regulated

voltage terminal block and the co.r=on terminal block.

The other lead line and the common line are connected to

the unregulated terminal block. From these three terminal

blacks comes all the power for the rotating column elec-

trical system, with the exception of the water cooler and

the lights inside the column mounting cabinet. An over-

load or power failure turns off all the power until the

magnetic switch is reset. The only power taken from the

unregulated terminal block was that used to operate the

vacuum pump. The right hand third of the control panel

was devoted to the control of power.


Heater Control

The heater circuit is shown on Page 30, Plate

XII. The power supplied to the side arm heater, to the

three column heaters, and to the kettle, is regulated by

five powerstat autotransforlers, 7 1/2 ampere capacity,










29

made by the Superior Electric Company of Bristol, Connecti-

cut. These transfomners are panel mounted in the horizon-

tal front portion of the panel. Each heater has an indi-

vidual fuse switch and pilot light, grouped on the sloping

portion of the panel. The voltage on any heater section

can be shown on a voltmeter by turning the voltmeter selec-

tor switch. The amperes passing through any heater section

can be read on an ximeter. The heater whose amperage is to

be indicated is selected by putting a jack in the correct

position. Plugging in the jack simultaneously disconnects

the usual connection to the common terminal block and

routes the current through the amoater. If more than one

jack is plugged in, the ammeter reads the cumulative

amperage. The heater wires are color coded, red being the

load line and white being the common line. All the power

line going- to the column are grouped together in a cable

which goes through a hole in the left side of the panel

into a hole in the right side of the column-supporting

cabinet. The cabinet serves as a support for the distri-

bution of the wiros. The heaters and the terminal con-

nections are described in the third chapter. Shorts .duo

to damage to the asbestos paper insulation on the sheet

metal tube were difficult to trac'ebecause of the inter-

connecting nature of the heater circuit.








PLATE XII


Unre ulated


Ter. Block
Power and Heater Circuits












PTAT.; XII

LEGIED



(= Fuse
S= S.P. TO-c Switch
= Pilot lirht
SP = Variac Pri a.ry
S Variac Secondnry
= Ar.motor
= Volt-actcr

Voltnetor Solector Switch
)-= Aoetor Soloctor Switch
1 = He aters



Operation of Anl Motnr J!-ck. Contacts B C normally closed.
Circuit is through 7 C B, Motcr out.When plug is
in.srted Contacts C B aro opo:ed, a::id circuit is
through D A B, iiotor in.












The rmocouplc System

Because of the critical nature of the adjustment

of the column heaters, an extensive thermocouple system

was installed to guide the control (Page 34, Plate XIII).

Eleven different thermiocouples may be selected by a 12

position rotary switch.. One position was not used. There

are four different functional groups of thermocouples, all

using the copper-constantin couple. One of these groups

is used in the control of the throughput and occupies

position twelve. This is discussed later under through-

put control. The other throe groups of thermocouples

are as follcAs: One group roads direct temperatures with

all couples using one ice-water reference cold junction.

This group occupies the one through five positions on

the thermocouple selector switch. The thornocouple in

position one reads the temperature in the distillation

head. The thermocouple in position two reads the tem-

perature of the upper column heater. The thermocouple

in position threo reads the temperature of the middle

column heater. The thermocouple in position four reads

the temperature in the lower heater section. The thermo-

couple in position five reads the temperature of the liquid

in the kettle. Those five thermocouples wore calibrated

at the boiling and freezing points of water and at

room temperature with these temperatures chocked by O\











Bureau of Standards calibrated thermometer, The calibration

curves are given on Plate XV, Page 38, Plate XVI, Page 39,

and Plate XVII, Page 40. The extension wires of these

thermocouples are in duplex insulation color coded, brown.

Another group of thermocouples are differential thermo-

couples which read the temperature difference between the

inside of the heater tube and the outside of the column,

These thermocouples occupy positions 6, 7, and 8 and

indicate the temperature differentials for the upper,

middle, and lower heater sections respectively. The ex-

tension wires for these couples are individual, and are

color coded, one blue and one black wire per couple. The

remaining group of thermocouples are also differential

thermocouples, but differ from the last group above in

that they indicate temperature differences between

different parts of the column. This group of thermo-

couples occupy positions 9 and 11. Position 9 indicates

the difference between the temperatures of the reflux

and the upper heater section. Position 11 indicates the

temperature difference between the kettle and the lower

heater section. The extension wires on those couples are

white. As shown in Plate XIII, page 34, all the thermo-

couples have a common line and a switch line. The common

line and the switch center tap, L, line are connected to

the potentiometer EM.F. Terminals.









PLATE XIII


TO T, T T3 To To To T7 To T,


A B C
Thermocouple System


T12


GA BA St.
cell


J 0














PLATE XIII

LEGE 1D


T,, T2, etc.


N

C

A


B and C

D



L

P.


Thermocouple in ic--water reference
cold junction

Thermocouples occupying positions one,
two, etc. as numbered on the switch

Constantin wire

Copper wire

Weston Standard cell, made by Weston
Electrical Instrumionts Corporation

1 1/2 volt dry cell batteries in series

Leeds and Northrup DC galvanometer,
No. 22811 b, Type iIS, sensitivity 0.05
microvolts per millimeter at 1 metor

Center tap line, rotary switch

Leeds and Northrup K-2 potentiometer











The optical system is shown in Plate XIV, page 37.

The light beam projector (A) is suspended from the top

of the panel and is powered by a 6 volt transformer (T).

The light beam goes from the projector to the prism (B),

to the galvanometer mirror (C), back to the prism (B),

up to the mirror (D) which is suspended from the panel

top, and to the back of the frosted glass scale (E)

mounted on sloping front of the panel. The light spot

is focused on the frosted glass scale by the lens (I).

Because of its location near machinery, shock mounting

of the galvanometer (F) became a necessity, The very

satisfactory and inexpensive method adopted consisted

of using a five-gallon can (H) filled with water and

mounted on four large rubber stoppers. The galvanometer

sat in a galvanometer and prism holder (G) Which in

turn rested on three rubber stoppers placed on top of

the five-gallon can.


Throughput Control

As will be shown in Chapter IV, the throughput

rate has a tremendous effect on the efficiency of the

column. Because of this, the control of throughput is

very important, particularly for analytical distillations.

The most commonly used device for controlling throughput

is the back pressure manometer, which measures the

difference in pressure between the still pot and the reflux

condenser. This device suffers from several disadvantages.












PLATE XIV


L- 15-" ---; 5" '- 12"
IE -32"
Optical System
The explanation for this drawing is on Page 36


31 I


50 "
I


-I-





PLATE XV
Thermocouple Calibration Copper-Constantin Couples in Positions 1-5
Temperature in Degrees Centigrade Against Millivolts


5


4









I-i




0 10 zO 30 40 50 60 70 80 90 100 ,
Temperature
Temperature o C.










PLATE XVI


38 .39 40 41 42 43

A E.M.F

Thermocouple Calibration, Temperature vs. Millivolts
per Degree


120

I O












PLATE XVII


1.2

I.I

1.0


.9

.8

.7

.6


.5

.4


.3

.2

.1


0 10 20 30 40


50 60 70 80


TEMPE RATURE, 0C
Thermocouple Calibration Correction to be Added to Tempera-
ture given in Table XV, Page 255 of Pyrometry by Wood and
Cork. McGraw-Hill Book Company, Inc. (1941) Second Edition


90 100











First the back pressure is not necessarily a function of

the rate of condensation in the head condenser or of the

rate of evaporation in the still pot. Under conditions

in which it is proportional, calibrations are necessary,

with tle values of throughput corresponding to the back

pressure readings established by independent means.

Obviously a change in a condition such as jacket tem-

perature could completely invalidate the calibration.

Diffusion of vapor into the high pressure line and any

change of its temperature are sources of operational

difficulties. Finally, the electric or photoelectric

relays used with the back pressure manometer provide

an off-on control with its accompanying oscillations in

contrast to the smoother continuous control which is

desirable. The throughput controller described herein

largely overcomes the disadvantages of the back pressure

manometer and has proved to be a valuable adjunct to the

rotating column.

This device controls the current to an internal

heater in the still pot according to the temperature

differential between the inlet and outlet lines of the

condenser. Ponsko, Quiggle, and Tonborg measured

the throughput rate by introducing thermocouples in the

inlet line as the cold junction. They used a copper coil

condenser and maintained constant water flow by means











of an elevated water tank with overflow to provide constant

head pressure. Glasobrook and Williams in "Technique of

Organic Chemistry" recommend "suitable baffles in the

condenser, to insure thorough mixing of the wator, and

proper insulation around the condenser to prevent heat

loss." However, in this laboratory, no baffles and in-

sulation were used with the accuracy being as good as

or better than the %j they reported. This was determined

by simple distillation of benzene, the hoat of vaporiza-

Stion of which is known, in such a manner that the entire

condensate went into a calibrated receiver. lWhen allowance

was made for the additional cooling of the condensate as

it ran down the condenser, the throughput was determined

as accurately as 1 or 2`. That insulation or vacuum

jacketing of the condenser is unnecessary is shown by

the fact that except in extremely slow water flow rates,

when no vapor was condensing, the temperature differential

usually became zero. Tests were run with various con-

densers and the temperature never was observed to exceed

0.10. except when the condenser water was so cold that

water in the air condensed on the outside of the condenser.

Of course with very low values of throughput those errors

become magnified so that other methods of measuring

throughput, such as counting drops from a calibrated

drip tip, become more suitable.











Differential thermocouples in the condenser lines

can be used to control throughput by means of a commercial

potentiomoter controller with a sensitivity of about 4.

microvolts or less. However, those devices are expensive,

especially if proportioning rather than off-on control is

obtained. The circuit shown in Plate XVIII, page .5t, was

constructed in this laboratory with a relatively modest

capital outlay and provides the desirable continuous con-

trol.

This continuous regulation was achieved by modi-

fication of a phase shifting A.C. bridge controlled thyra-
(1)
tron circuit originally published by Benedict for con-

trol of furnaces and thermostats, and later improved by

Tarnopol for control of annealing furnace temperatures.

As originally designed, the circuit employed a

resistance thermometer as the temperature sensing element,

but, in order to increase the sensitivity and speed of

response and lower the size and heat capacity of the

sensitive unit, thermistors wore employed. Since thermi-

stors have a negative temperature coefficient of resistance,

the thermistor in the condenser out line (RT) must be

connected to the same arm of the bridge as the controlling

decade resistance (Rx). Also, the thermistor in the

inlet line, (RB) which compensates for the temperature of

the incoming water, must replaced in the same position in











the bridr:e as the former resistance thermometer. In order

to compensate for the extra resistance because of the

thernistor in the decade arm, a 30-60 oln resistor (R.)

must be added to the former resistance thernmooeter arm

of the bridge. The modified circuit is shown in Plate

XVIII, Pago 45. The thermistors used in this laboratory

were obtained from the Frieze Instrumont Division of

Bendix Aviation Corporation in Baltimore, Maryland, and

were of the rod t-pe with a resistance of about 57 ohns

at 200C. and a temperature coefficient of resistance of

-2.0% per C. at 20C. The thermistor and lead wires were

waterproofed with a film of silicone Crease, and placed

directly in the water stream in the condenser (Page 39,

Plate XVI).

The thyrotron will safely pass about 2.5 anps,

and since the sensitivity of the circuit is greatest when

the thyrqtron is passing about one-quarter cycle, the

resistance R, in the anode circuit should be set so that

the maximum current that the tube will pass is about

twice that required to maintain the desired throughput

and in no case greater than 2.- amps. In practice the

external uncontrolled heater is set so thrt about one

amp, is required to maintain the desired throughput and

rheostat R5 adjusted to allow a maximum of about two

amps. The control point of the throughput is set by

increasing the resistance of the decade Rx until the











PLATE XVIII


Thyrqtron Temperature Differential
Throughput Control Circuit











PLATE XVIII

LEGEITD

RT = Thermistor in condenser outlet

RB = Thermistor in condenser inlet

R = 50,000 ohms potentiometer

RI = 10,000 ohms, 1 watt

Rs = 25,000 ohms, 1 watt

R, = 10,000 ohms, potentiometer

R4 = 5ooo ohims, 1 watt

RB = 0.5 megoihm, 1 watt

Rc = 10,000 ohms

R, = 10,000 ohms

Re = 50 olns, precision

Rx = 111.1 ohms, decade box

Rio = 25,000 ohms

Rt, = 31. ohms variable resistance

Rz = 30 ohms, 1 watt
R,2 = 1 megohm

RT = 20 dms Pt wire

C = 8 microfarad dry electrolytic, )150 volts

C, = 10 microfarad

C2 = 0.5 microfarad

C = l- microfarad, paper

CA = 0.02 microfarad, approximately












PLATE XVIII

LEGENTD (Continued)


C, = 1 microfarad, mica

L = 12 henrys, 75 ma

L, = 1000 henrys

T = Thordarson transforeor T 61 F 85: 5 volts secondary
center tap grounded

T, = Thordarson transformer T 60 F 94: 2 1/2 volts
secondary, center tap grounded

T2 = Thordarson transformor T 70 R 21 : power transformer
350 volts each side center tap, 70 ma; 6.3 volts
3 amp; 2.5 volts I4 amp.; 5 volts 2 amp.
T3 = Thordarson trrssformer T 30 A 20 : ratio: 1:2


A = Ammeter











desired throughput is attained. To obtain maximum sensi-

tivity, the potentiometer R is set so as to obtain the

greatest variation of anode current with variations of the

decade box resistance R,, without the control becoming

off-on. Since the position of the sensitivity control, R,

affects the control point, it should not be changed once

the control point is set by Rx. The control point was

found to be quite stable. Throughputs up to five liters

per hour wore maintained for days with loss than 5i varia-

tion. After shutting off the heater switches and allowing

the still pot charge to become cold, the throughput rate

could be quickly reestablished at the same value by turning

the switches on.

The throughputs were measured by thermocouples

also in the condenser as shown in Plate XIX, Page 51.

There differential thermocouples occupy position 12 on

the rotary switch (S) in Plate XIII, Page 34, and are

numbered T,,. The feed-back lag to the control s7stom

from ordinary condenser systems was entirely too slow.

Two designs wore evolved to increase the feed-back rate.

The simplest is to reverse the water flow by running

water in at the top. This required the outlet line

from the bottom to bo held above the inlet line to

keep the condenser filled. By thus having the water flow

countercurrent to the vapor, the time lag while the water











ascends the condenser is avoided. A superior method of

increasing food back rate employed is the high speed con-

denser. This condenser (PC. d1, P1. XIX and Pg. 52, Pt.

A in P1. XX) was fabricated by selecting two pieces of

glass tubing with small annular clearance, and keeping

them centered by wrapping a cardboard sheet around the

smaller one and sliding it into the larger one. After the

condenser is completed the cardboard is dissolved out with

sulfuric acid. The advantage of keeping the water jacket

volume small is seen by noting that the time required for

passage of the water from one end of the condenser to the

other is the jacket volume divided by the rate of water

flow.

The problem of constant water flow proved nore

difficult than was anticipated. The cornon centrifugal

type of circulating pump proved to be completely unsatis-

factory. An inconvenient alternative was the aforementioned

elevated constant pressure head tank. A very satisfactory

solution (Pg. 52, P1. XX) involved recirculating the

condenser water by pumping it through 3/8 inch copper

lines from an insulated tanh, in which it was cooled to

from 10 to 200C. Thi temperature was controlled by a

commercial constant temperature refrigeration device

(Part R). A rotary vane type of positive pressure pump

was used to circulate the water. The pump is turned on









50
by switch (S). Pilot light (P) lights when the switch is

closed. The water flow was measured and regulated by

first passing it through a rotameter with a built-in

valve. The voltage applied to the pump was regulated by

a variable autotransformer so as not to build up too

large a pressure behind the valve and overload the pump

motor. A by-pass valve was not used because it inter-

fered with regulation. Water flow was maintained at from

300-1100 ml./minute depending on the magnitude of the
throughput rate.








PLATE XIX


K'

A~


Inlet Side Arm of High Speed Condenser


r







PLATE XX




io
A

B






GH __


Water System












PLATE XX

LEGEND

A Main Condenser, High Speed

B Product Condenser

C Vapor Dividing Distillation Head

D Upper Thermistor Lead Wires

E Lower Thermistor Lead Wires

P Thermocouple Lead Wires

G Flowrotor, 0 to 1200 ml. per minute, Fischer and
Porter Company

H Rotary Vane Positive Pressure Pump, Eastern Industries
Model VW 1

I Inlet line, Copper, 3/8"

J Outline line, Copper, 3/8"

K Thermoregulator, American Instrument Company

L Line from Refrigeration Unit to Cooling Coils

H Lead wires from Refrigeration Unit to Thermoregulator

N 110 volts, 60 cycle, unregulated

0 Variable Autotransformer

P Pilot Light

Q Cooling Coils

R Cold Water Tank, Porcelain Covered Steel, Insulated
with Styrofoam.

S Switch on Panel

T 110 Volts, 60 cycle, regulated

U Refrigeration Unit, American Instrument Company












Column Motor Assembly


The rotor was driven by a five-eighths horsepower

series wound motor made by the Black and Decker Manufac-

turing Company. This motor was capable of running up to

10,000 R.P.M. It drove the rotor with a V-bolt running

on a five-inch pulley on the column shaft, and a three

and one-half inch pulley on the motor shaft. This pulley

ratio causes the motor to run 1.43 times as fast as the

column. A previously used one-third horsepower motor

was able to run the column at a two to one ratio, but

was definitely overloaded. The Black and Decker motor

was capable of pulling the column at speeds in excess

of 6000 R.P.M. The motor circuit is shown on Plate XXI

on page 57. The motor spood could be continuously varied

up to the maximum speed by a powerstat variable auto-

transformer (I). The power for the poworstat was taken

from the regulated load Terminal Block (L) and the common

terminal block (M). The common terminal block connected

directly to the powerstat common terminal. The load line

ran through a relay (J), then to the main switch (G).

The fuse (P) protected the powerstat and motor (H) from

overloads. The pilot light (K) indicated when the relay

(J) was closed. The control leads from the relay were

connected to the U-tube (C) by mercury wells around

tungsten wires sealed through glass.










55
This U-tube was used as a safety device to protect

the column from running without liquid reflux which was

necessary to lubricate the bearings. This could be ac-

complished in two different ways.

First the U-tube could be used as a back pressure

manometer. Sulfuric acid cr other conductive fluids could

be used to complete the circuit between the two tungsten

electrodes in the U-tube. If the kettle line was connected

to stopcock (A), then the liquid level was brought to where

the liquid was below the upper tungsten electrode by the

desired safety factor. Then the relay was set to be

normally closed so that, as long as the back pressure was

maintained, the relay (J) remained closed, allowing the

power to go to the motor. If the kettle line was connected

to stopcock (B), the liquid level was adjusted so that

the desired back pressure kept the upper tungsten electrode

covered by the desired safety factor. When the liquid

level in the left hand tube dropped due to reduced back

pressure, the relay cut the power to the motor. The relay

in this case was normally open.

The second way in which the U-tube was used as a

safety device was to connect both stopcocks (A) and (B)

to the vacuum line and when the operating pressure was

reached, one of the stopcocks was shut off. The relay

then was set normally open or normally closed depending










whether stopcock (B) or stopcock (A) respectively was

turned off. This second system of using the U-tube

safety device cut off the motor when the pressure went

above a certain value. Since a few millimeters pressure

increase could almost completely stop vaporization in the

kettle, this mode of operation was quite important if

the column was to be left unattended for any length of

time. When operation of the safety U-tube was not desired,

the relay could be manually opened or closed by means

of the switch (D). The throughput controller, previously

mentioned in this chapter, page 36 was also a valuable

protection to the bearings. For example, it kept the

throughput constant during transitions from the boiling

point of one fraction to a higher boiling fraction or it

could provide the pot with extra heat if a heat loss

developed anywhere in the column.

Tachometer

The revolutions per minute of the rotor were

indicated by a Weston A. C. Voltmeter calibrated for

R.P.M. The meter, located on the sloping front portion

of the control panel, read from 0 to 10,000 R.P.M. in

units of 100 R.P.M. The A. C, Potential, proportional

to the R.P.M. was generated by a Weston A.C, tachometer

generator (Pg. 59 P1.XXIJ, This generator was mounted

above the top of the rotor shaft and was connected to it








PLATE XXI, a



H


E
p --




J


L M

Motor Circuit

PLATE XXI, b
T


Timing Circuit


Q R











with a piece of heavy rubber tubing. The top three-fourths

inch of the column was reduced to one-half inch diameter to

allow the rubber tubing to be attached. A formerly used

tachometer drive consisted of a rubber edged pulley on the

generator which was driven by the outside edge of the

column V-belt. This arrangement worked quite satisfactorily,

but required calibration, and was changed to the present

system when the new shaft was installed. The tachometer

lines were color-coded rod,


Vacuum System

A drawing of the vacuum system is presented on

page 62, Plate XXIII. The rotating shaft seal, which was

the major problem of vacuum operations, was discussed in

Chapter II. The pressure in the system was lowered by the

Cenco Hy-Vac pump (V), which was connected by throe-eighths

inch copper tubing (Z), to the column (0), to a surge

bottle (N) and to the control panel. The rest of the

vacuum lines were tygon tubing. The pump was turned on

by a switch on the panel. A pilot light beside the

switch indicated when the pump was in operation. The

pressure in the system was regulated by a sulfuric acid-

diethylene glycol manostat (U) whose control lead wires

go to the relay (R). The relay bled air into the vacuum

system whenever the manostat fluid completed the control

circuit (S). When the relay was bleeding air into the














YzA TzX Xi


Moter Aasaebly
and
Head Assembly











system it turned off the pilot light (P) on the control

panel beside the manostat switch. Thus satisfactory

operation of the vacuum system was indicated by the

blinking of the pilot light. Wired in parallel with the

manostat was a light which illuminated the manostat tube

(U), indicating its operation. The pressure in the system

could be read in units of 0.1 millimeters, up to 100

millimeters, by a Zimmerli gauge (B), (not shown). A

larger Zimmerli type gauge reads up to atmospheric pres-

sure. The vacuum pump was protected from vapors escaping

the condensers, by a cold finger dry ice trap (L) which

has a large capacity for trapped liquid. This trap permitted

convenient visual observation of the amount of material

being trapped. It was easily cleaned out since the flask

which collected the liquid was attached to the cold finger

and vacuum line by standard tapers. The motor safety

U-tube (A) was discussed in connection with the motor

assembly on page 54. In this diagram the vacuum line is

connected to the U-tube to allow shutting off the motor

when the pressure in the column increases. Back pressure

is determined by a U-tube .(T), one leg of which is con-

nected to the vacuum system and the other leg of which is

connected to the kettle. Considerable difficulty was

encountered in attempting to prevent vapor diffusion from

the kettle into the back pressure line (Y). When vapor

diffused into the line it condensed and soon the line











began to fill with liquid. The best arrangement found,

was to use as the back pressure line from the kettle, a

large diameter long glass tube (Y') with a large stopcock

close to the kettle. This long tube acted as a vapor

interface chamber and as a condenser. It had sufficient

volume to register the correct value of the back pressure

over the range of pressures encountered. This range was

from 0 to 25 mm. fg. Further diffusion into this line

was retarded by a capillary tube (X) and the small rubber

tubing (Y) used as the back pressure line to the control

panel.


Distillation Head

A satisfactory head for the rotating column

must: (1) allow high throughput rates without flooding,

(2) be able to maintain a constant reflux ratio in the

range 1:1 up to 200:1 and (3) permit no leakage of

product into the receiver when on total reflux. All

these requirements were met by modification of a vapor

dividing head described by Collins and Lantz. 3

The modifications which resulted in the head are shown

on Plate XXIV, page 67, and are as follows. The vapor

passage was enlarged to a minimum of 0.5 inches cross

section area. (Throughout the metal portion of the column,

a minimum vapor passage cross sectional area of 0.75

inches was maintained, except for the annular space










PLATE XXIII


Vacuum System










63


PLATE XXIII

LEGEND


A Motor Control Safety U-tube

B Zimmerli Gauge

C Main Condenser

D Product Condenser

E Termocouple Lead Wires

F Thermistor Lead Wires

G Motor Control Safety Relay

H Switch for Manual Relay Operation

I Power Lines from Regulated Terminal to Relay
Switch 60 Cycle, 110 Volts

J Power for Relay Operation, 60 Cycle, 110 Volts

K Power Lines to Motor Circuit

L Dry Ice Trap

M Distillation Head, Vapor Dividing

N Surge Tank

0 Rotating Concentric Tube Distillation Column

P Pilot Light

R Manostat Air Bleeding Relay

S Manostat Control Lines

T Back Pressure Manometer, Mercury

U Manostat Control Tube, Sulfuric Acid-Diethylene
Glycol











PLATE XXIII

LEGEND (Continued)


V Vacuum Pump, Cenco IIy-Vac

W Kettle

X Glass Capillary Tube

Y Back Pressure Line, Rubber, 3/32 I.D.

Y Vapor Interface Chamber

Z Copper Tubing, 3/8 inches











between the rotor and the outer tubes. This area was

about 0.41 square inches.) For safety, and protection

from vibration, a 50/30 semi-ball joint was used to

connect the head to the column. The product condenser

was connected with a 35/20 semi-ball joint. The con-

denser was attached with a 3/415 standard taper. A ground

glass taper for a standard taper thermometer was provided

in the side arm, but it was actually fitted with a thermo-

couple well. Expansion bellows were made in the inside

structure as well as in the vacuum jacket. The product

condenser and receiver (Plate XXII, page 59, Plate

XXIII, page 62) were altered. This was to allow increasing

the condensing capacity by use of an Allyn condenser and

to permit use of receivers which could be changed without

destroying the vacuum in the column. Most of these

modifications were made as a result of tests with an

experimental head. This experimental head was made from

the original Collins and Lantz design for the purpose of

developing a head for the rotating column. The most

serious difficulty encountered with this experimental

head was leakage of liquid reflux into the product line.

This was caused by reflux dripping onto the ground glass

ball valve and either leaking around the grinding or

dropping into the product line when the valve was opened.

In a vapor dividing head such leakage of liquid can greatly









affect the reflux ratio and make it difficult to maintain

total reflux. This leakage was eliminated in the present

head by use of a female 28/15 semi-ball joint as the

valve plunger. This female semi-ball joint was seated on

a male 28/15 semi-ball joint as shown in Plate XXIV, b,

page 67. Leakage between the two ground surfaces was

retarded because of gravity, and the greater area of

contact between the two parts of the ball joint. Leakage

of drops, clinging to the plunger, into the product line,

was prevented because the female semi-ball joint acted

as an,umbrella over the product line port. An advantage

in the fabrication of this head was that these standard

semi-ball joints are commercially available and did not

require hand grinding. The 28/15 female semi-ball joint

did not quite pass through the 34/45 female standard

taper at the top of the head, and had to have soxm of the

edge removed by grinding. Thus when the male 34/45

standard taper on the condenser was inserted, it provided

a seat for the plunger when it was lifted by the solenoid

at the top of the condenser. This shut off the main

condenser at the same time the product condenser was opened.


Timing Circuit

The function of the timing circuit (Plate XXI, b,

page 57) was to provide the solenoid at the top of the
main condenser with intermittant power.









PLATE XXIV, a


Automatic Vapor Dividing Head

PLATE XXIV, b


S-STANDARD TAPER
40 /50
EDGE GROUND
SEMI-BALL29


Details of Vapor Dividing Valve


--77-









68
The reflux ratio is theoretically the ratio of

the time the power is on, to the time it is off. Thus

it is important to control, not only the frequency of

application of power to the solenoid, but also the duration

of the power-on period. Actually, the reflux ratio will

vary from the time ratio somewhat, depending on the

particular head to which the power is supplied. In the

present head the actual reflux ratio tends to be higher

than the time ratio.

A device which switches power alternately off and
on is called a repeat cycle timer. After several types

of repeat cycle timers proved unsatisfactory, a suitable

timer was obtained from the G. C. Wilson and Company,

Huntington, West Virginia. The company supplied the

timer with the time limits of the usual off-on cycles

reversed. This modified model 1 timer has continuously

variable cycles, with the on cycle approximately from

0.1 to 60 seconds and the off cycle approximately from

0.1 to 300 seconds. As shown on Plate XXI, page 57,

the 110 volt, 60 cycle, power from the terminal block

roes first to the repeat cycle timer (X) where it passes

through a double pole, double throw relay switch. Power

is alternately switched from the on cycle to the off

cycle. Although only the on cycle power is used for the

lifting solenoid, power from both cycles is used for the

timer (T) so that their duration may be measured. This









69
timer operated on the principle of charging and discharging

condensers through variable resistances. The throe-way

switch (S ) permitted either the on cycle or the off cycle

to be timed. (Yi) was the load line for the on cycle.

(Zi) was the load line for the off cycle. (Y,) was the

common for both cycles. The switch (U) turned the timer

off and on and allowed the recepticle (R) to be isolated

from the repeat cycle timer. With the switch (U) off, the

recepticle (R) could be used to time another repeat cycle

timer. With the switch (U) on, the rocepticle (R) could

be used to provide timed power for another column. The

switch (11) connected the repeat cycle timer with the sole-

noid (Q) on top of the condenser. By use of the switch

(W), the time intervals could be adjusted to the desired

values before turning on the solenoid. (V) is the vari-

able autotransformer htich was used to adjust the voltage

across the solenoid terminals. The voltage was adjusted

so as to obtain positive lifting of the valve without

damaging it or the condenser standard taper, against which

it seats. A second recepticle (RI) permits the solenoid

load wire to be easily disconnected. The off-on switch

for the repeat cycle timer, the hifh-low range switches

for the cycles, the potentiometer rotary switches, and the

synchronizing switch, are all part of the commercial repeat

cycle tirer circuit and are not shown in Plate XXI.














CHAPTER IV


TESTING AID EVALUATIOi OF THE ROTATING COLUMN


The main objectives in testing the column were set

forth in Chapter I. The experimental work outlined in this

chapter was planned to give a general picture of the method

of operating the rotating column, and the interrelationship

of the variables which determine the operating characteris-

tics. The variables which could be directly controlled,

other than construction variables, are as follows: (1) The

pressure of the system (measured at the condenser and here-

after referred to as condenser pressure) which,when opera-

ting at reduced pressure, was controlled by a manostat and

measured by a Zirn~erli gauge. (2) The heat input to the

colt-n heaters was controlled by a poworstat and measured

by a voltmeter and an ammeter. (3) The heat input to the

kettle external heaters was controlled by a poworstat and

measured by a voltmeter and an ammeter. (Lj) The heat in-

put to the pot internal heater (when used) was controlled

by the thyratron throughput controller. (5) The power

supplied to the motor which drives the rotor was controlled

by a powerstat. These controls were discussed in detail in

Chapter III. The dependent variables which were affected









71

by some or all of the above directly controllable variables,

are discusscd below in torms of the variables which affect

them and of the method used to measure them,


(1) Ono of the more important variables was the

throughput. There are several definitions of "throughput."

One states that the "throughput boilup rate, or vapor velo-

city, is the rate at which the vapor is passing up the

column and is usually expressed in terms of the quantity

of liquid equivalent to the vapor passing up tho column per

unit time."(17) Another states, "The vapor velocity may be

expressed either as unit weight per unit time, of material

reaching the top of the column, or as unit distance per

unit time." (2 ) These definitions may be consistent and

sufficient if the column is being operated under adiabatic

conditions. However, when adiabatic conditions are not

maintained, the various terms quoted above do not have the

same meaning as the boilup rate would not be the suae as

the condensation rate at the head condenser.

To describe clearly non-adiabatic operating con-

ditions, both the boilup rate and the condensation rate

at the head should be stated.

In this investigation the term "throughput" was

arbitrarily defined as the moles of material per unit time

passing, up the column when the operating conditions of the











column were adjusted so that the boilup rate was equal to

the condensation rate at the head condenser. This defini-

tion seems to give the most consistent correlation with

the other variables. The definition of throughput is of

particular importance in describing the operation of the

rotating column for two reasons. First, the throughput

has a critical effect on the efficiency and other charac-

teristics of the column. Secondly, the rotating concentric

tube column differs from all previously reported types of

distillation columns in that the rotation produces a largo

amount of heat in the column. The amount of heat released

by the rotation depends on the viscosity and quantity of

the material passing through the column as well as the

friction in the bearings. The heat produced by those

effects was equivalent approximately to an increase in

condensation rate at the head of 500 ml. per hour of the

test mixture per 1000 R.P.i. of rotation. For example,

at 5000 R.P.!:., with a boilup rate of 500 ml./hour at the

kettle, over 2000 ml./hour would be condensing at the

head if the column heating jackets were maintained at the

boiling temperature of the reflux. Thus the conventional

methods of operating a column adiabatically by surrounding

the column with vacuum jackets, heavy insulation, or air

heated to the column temperature, results in wide deviation

from adiabaticity in the rotating column.










73
In practice the throughput was controlled ap follows:

First, the heat loss in the kettle charged with distillant

was determined. This was done by first plotting the kettle

temperature against the ampores passing through the pot

heater to produce that temperature. An example of the re-

sulting curve is shown on Plate XXV, page 74. The hori-

zontal portion of the curve at the top indicates that the

boiling point was reached and that additional heat went

into vaporization of the charge. The wattage of the heat

loss was calculated by multiplying the amperes at the break

in the curve by the voltage at that amporago. This wattage

was then added to the wattage required to vaporize the

material in the still at the desired rate. This latter

wattage was calculated from the heat of vaporization at

the boiling point. The heats of vaporization used in this

investigation were as follows:(4) N-noptane at the boiling

point has a heat of vaporization of 75.6 calories per gram

and at 250C. a heat of vaporization of 87.2 calories per

gram. Methyl cyclohcxane has a heat of vaporization at

the boiling point of 77.2 calories per gram and at 250C.

of 86.1. Between these temperatures a linear relationship

was assumed. The column heaters were then adjusted to

provide the same condensation rate in the head condenser

as the boilup rate in the kettle. At )4000 R.P.H. the

column heaters had to be operated at about 10C. below

the kettle temperature to keep the boilup rate and the










PLATE XXV


4.35


(I)
.0
O


4.1
4.15


1.05 1.10 IU5


120 125 130 L35 140 145 1.50 155


AMPERES


Kettle Amperes vs. Thermocouple E.M.F. in Millivolts









75

condensation rate the same. Thus at higher values of

R.P.M. the heating jackets served essentially as a con-

trolled heat leak to remove the heat generated by rota-

tion. Once the column jacket temperatures had been ad-

justed for a particular R.P.I., the throughput could be

controlled by adding or subtracting the calculated wat-

tage from the kettle heater. This indicates that

the overall heat absorbed or lost in the column was re-

duced to a very low value, and that the column was opera-

ted under conditions which simulated adiabatic conditions

in so far as the liquid and vapor were concerned.

(2) The revolutions per minute of the rotor do-

ponds on the power applied to the driving motor, the

frictional drag of the liquid and vapor on the rotor,

and the bearing friction. At low throughputs the R.P.M,

was sometimes increased by an increase in throughput,

presumably because of improved bearing lubrication. How-

ever, at higher values of throughput, an increase in

throughput slowed the rotation. For example, an increase

from 1 to 2 liters per hour, usually lowered the R.P.i.

by 200 or 300 R.P.ei.

(3) The pressure drop in the rotating column is
principally related to the R.P.II. and the throughput.

These relationships are shown on Plate XXXV, page llQ and

Plate XXXV3 page 112. The pressure drop was measured by a

mercury U tube.











The pressure drop in the rotating column did not

give a very satisfactory indication of throughput because

of its very low values, the difficulty of measurement, and

the fluctuations (even though the throughput as indicated

by thermocouples remained constant). Since the pressure

drop was affected by more than one variable, it.is diffi-

cult to attach any useful significance to it. The principal

value of the pressure drop measurement in the rotating

column was in indicating floods. When the pressure drop

exceeded about 25 rmn., a flood soon became evident at tihe

head.

(4) The theoretical plates in the rotating column
depend principally on the throughput, the R.P.N, and the

condenser pressure. The definition of throughput given

on page 71 was selected particularly because it clearly

designates the quantity of material passing through the

column. Hence it permits a satisfactory correlation between

throughput and platage. This relationship is shown on

Plate XXX1D, page 107, and Plate XX~IV, page 10 The highost

possible theoretical platage nay not necessarily be ob-

tained by operating the column according to the definition

given above. If the column were operated according to the

definition, and one of the rates lowered with the other rate

held constant, there were indications that the number of

theoretical plates went up. Also, there were indications

that raising either the boilup rate or the condensation












rate with the other rate held constant, caused a decrease

in the platage. Although having a lower boiling rate than

condensation rate gives a higher platage at total reflux

than if the rates were equal, at the higher value the effect

on an actual fractionation where the reflux ratio is finite

may or may not be advantageous. The lower boilup rate

probably would-not increase the sharpness of separation to

the same extent as indicated at total reflux, This is

because with the lower boiling rate the column would equi-

librate more slowly, and thus the separation would be more

adversely affected by a change to a finite roflux ratio

than it would be if the boilup rate were the same as the

higher condensation rate. The beneficial effect of the

rotation is shown on Plate XXX~O page 107. The detrimental

effect of the pressure is shown on Plate XXIX, page 124.

(5) The operating and static holdup of the rota-

ting column were determined at atmospheric pressure. By

operating or total holdup is meant the quantity of material

in the column during distillation. Static or nondrainablo

holdup is tih quantity of material remaining in the column

after the distillation is shut off.

Several methods of determining holdup are discussed
(17)
by A. and E. Rose in Distillation. The present method
is a modification of a method described by Tonberg, Quigr3











and Fenske. ) This method makes use of the change in

composition of a non-volatile solute in the volatile liquid

whose holdup is to be measured.

Two improvements wore made on the original method.

First a new test mixture was developed. N-heptane was used

as the volatile component because it was the more volatile

component of the platage test mixture, and because it had a

suitable refractive index. The non-volatile component was

dibutyl sobacate which has a boiling point of 2500C. at

100 r;.m of IHg pressure. The di-butyl sebacate used was

vacuum distilled and had a refractive index of 1.4398.

Since there is a refractive index difference of 0.0546

refractive index units, the mixture is easily analyzed by

the refractometer permitting the use of very small samples.

The use of small samples permits several determinations of

holdup to be made under different conditions without cor-

rections for sample size. A. and E. Rose(17) report the

use of rosin oil which permitted analyses by refractive

index but since the rosin oil varies in composition from

batch to batch,'the relation btweeon refractive index and

composition must be determined for coch batch.

The data for the relation between refractive index

and composition, both volume percent and weight percent,

is given in Table I and is plotted on Plate XXVI, page 81.

Because of the difficulty in accurately rmasuring volumes











of dibutyl sobacate, the weight percentage data is more

accurate and was used in the present determination of hold-

up. The use of a liquid non-volatile component prevents

possible deposition of solid material at high concentrations

and resulting carbonization.

The second improvement made in the determination of

holdup was the sampling method. The flask used for the

holdup measurements is shown on Plate XXVII, page 83 ,

The sampling line reached almost to the bottom of the Ikttle.

A rubber tube with a rubber bulb was attached to the side

tube on the chamber at the other end of the sampling tube.

The sampling tube extended almost to the bottom of the

calibrated receiver. This receiving tube was attached to

the chamber by a 19/38 standard taper. To remove a sample

for refractometric analysis, the stopcock in the sample

line was opened and suction was applied by the rubber bulb.

When the sample tube was fill] d, pressure was applied by

the rubber bulb, to force the test mixture back into the

kettle. This procedure was repeated several times so as

to obtain a completely representative sample. The long

tube to the bottom of the receiver permitted forcing all

but the few drops needed for rofractometric analysis back

into the test kettle. After the throughput was adjusted,

at least one hour was allowed for equilibrium to be

established before taking samples.







DATA FOR THEi


R:LTION
MIXTURE


TABLE I
BETI~EE E COMPOSITION AND REFRACTIVE IIDEX OF THE
I1-IiPTAIE AND DIBUTYL SEBACATE


Wt. of Wt. of I- Total Weight Volume of Volume Volume 20
Dibutyl Hoptano Weight Percent Dibutyl of Percent nD
Sebacate Grams Grams Dibutyl Sebacate 1i-heptane Dibutyl
Grams Sebacate 2ll. l. Sebacate

16.908 0 25 0.0 1.3852
2.319 15.251 17.570 15.21 2.50 22.50 10 1.3912
1,.677 13.548 18.225 25.6I4 5.00 20.00 20 1.3971
6.964 11.941 18.905 36.86 7.50 17.50 30 1.4L028
9.368 10.179 19.5547 4.7.92 10.00 15.00 40 1.L084
11.601 8.567 20.168 57.I46 12.50 12.50 50 1.4140
13.924 6.798 20.722 67.20 15.00 10.00 60 1. 193
16.133 5.211 21.347 75.55 17.50 7.50 70 1. 4241
18.538 3. t1 21.979 84.35 20.00 5.00 80 1.4294
20.843 1.754 22.597 92.24 22.50 2.50 90 1.4343
25 0 100 1.4398






PLATE XXVI


1.39


25
"n D
Percent Dibutyl Sebacate Against Refractive Index
A Voltune per cent; B Weight per cent


100



90



80


70



60


50



40


30












The holdup was determined as follows. About

250 ml. of approximately 30 volume percent dibutyl sebacate

was placed in the weighed holdup flask. This concentration

was selected because the higher concentrations give a

greater change in concentration for a given holdup. Two

hundred fifty milliliters of test mixture were used so that

the sample taken would be negligible and enough n-heptane

would be available for the expected range of holdup. The

refractive index of the test mixture was taken, the cor-

responding percentage road from the graph on Plato XXVI,

page 81. This percentage multiplied by the weight of the

test mixture gave the weight of dibutyl sebacate in the

kettle. After the throughput equilibrium was established,

a sample was taken as described above and the refractive

index determined. The corresponding weight percentage of

dibutyl sobacate was read from the graph. This data per-

mitted setting up the following material balance:


Weight fraction Weight of dibutyl sebacate in
in sample original mixture
IWt. of original mixture holdup


For example, the original mixture weighed 19h.5

grams containing 69.4 grams of dibutyl sebacate and at a

6.7 moles per hour throughput, the weight percent was 85.8.










PLATES XXVII


Pl*tUi' of Ho14up Appexatus











This gives:


0.858 = 69.4.
194.5 x

69.41 = 166.9 0.858 x

97.5 = 0.858 x
x = 114 grams

The molecular weight of n-heptane is 100.2

14 = 1.1l moles holdup at 6.7 moles per hour and 0 R.P.M.
100

The holdup data is given in Table II, page 85. The
holdup was also determined when the column was rotating at

2000 and 4000 R.P.M. Rotation was found to greatly reduce

holdup but the speed of rotation seemed to have little effect,

2000 and 4000 R.P.M. giving almost identical values. The
holdup at 0 throughput is the static holdup and seems to

account for a large portion of the operating holdup. This

is probably partly due to the cross arm. Because of space

limitations, the flexible joint in the cross arm had to be

mounted almost horizontally and this did not drain as well

as if it were mounted vertically.











TABLE II


HOLDUP OF ROTATING COLUMN
TEST MIXTURE IH-HEIPTANE
DIBUTYL SEBACATE

Through ut p 20 Weight / Holdup Moles
Mole s/Ir. bl .. nD Dibutyl n-hept&ne
Sebacato

3.4 0 1.4252 77.5 1.05
6.7 0 1. 3 03 85.8 1.14

13,4 0 1.4351 93.5 1.20
6.7 0 1.4303
chock
6,7 4L000 1.4213 70.7 0.964
6.7 2000 1.4206 69.5 0.947
0.0 0 1.4126 55.0 .691















PLATE XXVIII


.8


6 --

4-



D


8

6

4



^-l l i


2,

B.

i.

i.

i.

Ij


Rotating Column Holdup


10 1 2 14 16
MOLES PER H


)


2 4 6 8
T HROUG H PUT,


Effect of Throughput on


18 20
OUR










Tost An7aratus nnd Sanpling


The vapor dividing head described in Chapter III

proved to be very satisfactory for withdrawal of small

samples of distillate. The samples were collected in 10 ml.

graduated tribes by applying power to the left valve solenoid

for a sufficient time (measured by the timer) to obtain a

sample of the desired size, generally about one milliliter.

The test kettle with accessories is shown on Plate

XXIX, page 89. It was equipped with an internal heater (A),

which was powered by the throughput controller described in

Chapter III, and which was connected to the three-liter,

three-neckld, round-bottom blask by a 28/15 semi-ball joint.

The back pressure manometer line was connected to

tube (C) with a 28/1 semi-ball joint. The tube (C)'could

be closed by a stopcock. The bores of the tube and the

stopcock wore large enough to allow drainage of the test

mixture which condensed in the larger diameter, horizontal,

vapor interface chamber.

Part (D) is a combination kettle sampling device

and thermocouple well. Di-butyl phthalato was used in the

thermocouple well as a heat exchange medium. The tempera-

ture in the well was indicated by the thermocouple in

position five on the thermocouple switch. The side tube,











which was used to withdraw kettle samples, was made from

thick walled capillary tubing to provide strength and to

reduce holdup of liquid in thoetube. When samples were

not being withdrawn, the smapling tube was shut off by the

stopcock (II). The sample was collected in a graduated

centrifuge tube (F). This receiver was connected to the

sampling tube by a 19/38 standard taper attached to the

splash chamber (E). This splash chamber was of sufficient

size to prevent the liquid being sampled from splashing or

boiling into the vacuum line.

A kettle sample was withdrawn as follows. The two-

way stopcock (G) had one arm connected to a pressure suf-

ficiently lower than the system to cause the liquid to be

forced into the sample receiver. The other arm was left

open to the atmosphere. The sources of the lower pressure

were a vacuum pump when the condenser pressures were less

than 50 mm. of Hg, and a water aspirator when the pressure

was greater than 50 mm. IIg. At atmospheric pressure a rubber

bulb was used. With the stopcock (G) opened to the reduced

pressure arm, and the receiver in place, the stopcock (II)

was opened to allow the kettle liquid to fill the receiver.

When the receiver was full, the stopcock (G) was not

adjusted, but the rubber bulb was used for alternate

suction and pressure. The liquid was forced in and out

of the sampling tube several times before a sample was

taken for analysis by use of the refractometer.










89

The purpose of the alternate filling and emptying

of the receiver was to assure a representative sample,

uncontaminated by stopcock lubricant, by small amounts of

impurities which might hove been in the receiver, or by

previous samples left in the sampling line and stopcock (H).

The tube reached almost to the bottom of the receiver and

thus made it possible to force all but a small amount of

liquid back into the kettle after the test lines and the

receiver were washed out with the larger samples.

This washing and sample size control is a feature

not found in previous kettle sampling devices. When using

other sampling devices, it is difficult to control the

sample size and to avoid contamination by the previous

sample and to return any excess sample to the kettle.




















F


Test Kettle with Vacuum Sampling Devico
Scale 1/2" = 1"












Operation of the Column and Test Procedure

A picture of the panel front with the controls
numbered is shown on Plate XXX, page 94. These controls

are discussed in detail in Chapter III. The first step

in operation of the column was to press the magnetic switch

button (35) which connected 110 volts to the terminal blocks.

The pilot light (33) indicated the position of the magnetic

switch. The heater switches (30) from left to right were for

the cross arm heater, top column heater, middle column

heater, bottom column heater,.and kettle heater respectively.

The power to the heaters was adjusted with the powerstats

numbered as follows: Powerstat number (31) controlled the
cross arm heater; Number (37) controlled the top column

heater; number (23), the middle column heater; number (32),

the bottom column heater. The power to the heaters was

adjusted first with the aid of the voltmeter (27) and the

ammeter (36). The voltage to be measured was selected by
the rotary switch. The positions on this switch were

numbered clockwise in the same order as the heater switches,

The various circuits could be connected to the ammeter by

the jacks (29) above the heater switches. The jack above
each switch selects the amperes passing through that switch.

When the column was almost up to the desired temperature,

the exact temperature was then obtained by the use of the




University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs