Quick viewing(Text Mode)

Visco Handbook THEORY and APPLICATION of VISCOMETRY with GLASS CAPILLARY VISCOMETERS Welcome to SI Analytics!

Visco handbook THEORY AND APPLICATION OF VISCOMETRY WITH GLASS CAPILLARY Welcome to SI Analytics!

We express our core competence, namely the production of analytical instruments, with our company name SI Analytics. SI also stands for the main products of our company: sensors and instruments.

As part of the history of SCHOTT® AG, SI Analytics has nearly 80 years experience in glass technology and in the development of analytical equipment. As always, our products are manufactured in Mainz with a high level of innovation and quality. Our electrodes, titrators and capillary viscometers will continue to be the right tools in any location where expertise in analytical measurement technology is required.

In 2011 SI Analytics became part of the listed company Xylem Inc., headquartered in Rye Brook / N.Y., USA. Xylem is a leading international provider of water solutions. We are pleased to introduce you to the Visco Handbook!

It replaces the previous brochure "Theory and Practice of Capillary Viscometry". This has been updated, redesigned and restructured. The information content of some areas such as analysis was increased and specially themed areas such as hydrodynamic principles were moved into their own appendices so as not to deter the reader by too much theory.

The focus was placed on the practical and general information needed by each user of capillary viscometry. Both experiences and policies of the relevant standards are considered here. We thus hope to provide you with a faithful companion for everyday laboratory work, one that you can use profitably.

We of SI Analytics are pleased to work with you successfully in the future.

SI Analytics GmbH

Dr. Robert Reining CEO CONTENTS

CHAPTER 1 - Rheology 1.1 Introduktion...... 7 1.2 Non-Newtonian flow behavior...... 11 1.3 Principles of viscosity measurement...... 13

CHAPTER 2 Fundamentals of capillary viscometry 2.1 Measurement principle...... 14 2.2 Designs of glass capillary viscometers...... 16

CHAPTER 3 Measurement of the flow time 3.1 Manual timing...... 19 3.2 Automatic timing...... 19

CHAPTER 4 Method for determining viscosity 4.1 Neglect of the kinetic energy correction...... 23 4.2 Application of the kinetic energy correction ...... 25 4.3 Experimental determination of individual kinetic energy correction (according to DIN 51562-3)...... 26 4.4 Examples of viscosity determination...... 28

CHAPTER 5 Calibration...... 29 CHAPTER 6 Handling of capillary viscometers 6.1 General guidelines for the selection of the measuring system...... 32 6.2 Cleaning of capillary viscometers...... 35 6.3 Preparation of the measurement...... 38 6.4 Performing the measurement...... 41

CHAPTER 7 Sources of error and special corrections 7.1 Correctable errors and corrections...... 47 7.2 Uncorrectable errors...... 51 7.3 Trouble shooting table...... 53

CHAPTER 8 Special application 8.1 Testing of plastics...... 57 8.2 Viscosity determination of oils and additives...... 68 8.3 Food testing...... 71

Appendix A Symbols and units used Bibliography Standards Authors edition:

Prof. Dr.-Ing. habil. Jürgen Wilke

Dr.-Ing. Holger Kryk

Dr.-Ing. Jutta Hartmann

Dieter Wagner

major rewrite, update and new edition (© 2015):

Dr. Andreas Eich Visco handbook CHAPTER 1 Viscosity characterizes the inter- nal of liquids and gases. VISCOSITY - RHEOLOGY If a medium is located bet- ween two plane-parallel plates, 1.1 Introduction it will require some amount of force to displace the upper plate. The subject of this Visco Hand- The liquid starts to flow inside book is viscometry with glass the gap. A layered flow builds up capillary viscometers. (Fig. 1). Viscometry deals with the de- y termination of viscosity and is F a branch of rheology, which is scientifically concerned with the v flow and deformation behavior x of materials. In rheology, sub- stances, especially viscoelastic Fig. 1 Basic shearing model in the ("semi-solid"), are examined, case of laminar, stationary flow which are classified as being be- tween a fluid and a solid, based on their mechanical properties. The fluid particles which are di- rectly adjacent to the plates are With glass capillary viscometers, firmly bonded to the surface by only the viscosity of samples with adhesion forces. In this process ideal flow behavior (Newtonian the fluid layer neighboring the liquids) is measured. This section plate being displaced adopts briefly explains the difference the velocity of the plate. All between ideal and non-ideal neighboring layers stay more flow properties. Additional infor- and more behind with the in- mation about rheology may be creasing distance to the plate found in [1,19]. being moved. The cause for this phenomenon can be found in cohesion forces which counter-act the reciprocal dislo- cation of the individual layers.

7 Visco handbook The shear stress τ refers to The relationship between dy- the quotient of force F and namic viscosity η and density ρ is the boundary surface A of referred to as kinematic viscosity the liquid: ν (pronounced "ny"):

F η 2 τ = ν = = m / s (1.5) A (1.1) ρ  

For reasons of convenience, The gratient of velocity, the shear the unit of mm2/s is used, rate γ is the differential quotient: which then corresponds nu- merically to the former unit. dv γ = (1.2) cSt (centistokes). dy

In Newtonian , the viscosity According to Newton's Viscosity remains constant with a change Law, the shear stress τ is propor- of the shear rate. As a conse- tional to the shear rate γ : quence of the differences in size, shape, and interaction between the molecules may change τ = η ⋅γ (1.3) η within very wide limits. The proportionality factor η (pro- nounced "eta") is referred to as Examples: dynamic viscosity. n-pentane 0.230 mPas (20 °C) The unit of viscosity is Pa · s, Water 1.002 mPas (20 °C) wherein low-viscosity samples Propanetriol 1480 mPas (20 °C) are usually specified in mPa · s (Glycerol) and thus the former unit cP (centipoise) numerically corre- sponds to:

τ 2 η =  Ns / m  =  Pa⋅ s  (1.4) γ    

8

γ Temperature dependence of This effect is of great practical im- viscosity portance, for example, in lubrica- tion technology, as will be shown During flow, liquid molecules later. slide against one another under the expenditure of energy (ac- In addition to the temperature, tivation energy). The proportion the pressure also has an effect of molecules that have this ener- on the viscosity: An increase in gy depends on the temperature pressure generally leads to in- and is described by the Boltz- creased viscosity. The effect does mann distribution. not appear so much in everyday life since only pressure increases This leads to the relationship: E of 10 to 100 bar (or even higher) - visk RT lead to a significant increase in η = k ⋅e (1.6) viscosity of liquids. The pressure dependence of the viscosity k Proportionality factor therefore plays no role in visco- metry with glass capillary visco- Evisk Activation energy of the viscous fluid meters, which is discussed in this R Gas constant Visco handbook. T Absolute temperature

After that, η strongly decreases for liquids with increasing tem- perature. As a rule of thumb, one can say that the higher the abso- lute values of the viscosity and the lower the temperature, the greater is the decrease.

9 Visco handbook Viscosity of mixtures/solutions A particular behavior can be ob- When liquids are mixed, the vis- served with the dependence of cosity can be approximately cal- concentration of viscosity of elec- culated from the of the trolyte solutions. individual components accord- ing to a logarithmic mixing rule. If the liquid layers are moving at different velocities, the deforma- Example of two components: tion of the ion cloud will cause the occurrence of additional inter- ionic interacting forces which will lnη = w lnη + w lnη (1.7) mix 1 1 2 2 affect friction between the indi- vidual layers. w1 , w2 : weight fractions of the components 1,2 H. Falkenhagen has derived the

limiting law of viscosity from the This rule only applies to mixtures theory of inter-ionic interactions of similar components, and does for highly dilute electrolyte solu- not generally apply for aqueous tions: solutions. For a better description of reality, equations must be η = η + K c (1.8) selected with adjustable para- c 0 meters [2]. ηc Viscosity for the ion The viscosity of the solutions of concentration c Viscosity of the pure solid substances is often higher η0 than that of the pure solvent. The solvent at the same specification is usually made as a temperature relative or specific viscosity (see K Constant which depends Chapter 8). on the following variables: - Temperature - Dielectric constant of the solvent - Ion valences - Ion mobilities

10 1.2 Non-Newtonian flow Examples: behavior - Lacquer/varnish - Thermoplastics Shear thickening (dilatancy) - Polymer melts Shear viscosity increases with ris- - Adhesives ing shear rate (work hardening: - Additives Fig. 2, Curve b). - Emulsions - Suspensions η Yielding point (plasticity) b The liquids only begin to flow at a minimum shear strain. Below this yielding point, the substance a behaves as a solid.

Examples: c - Paints, dispensions, cream - Food (mayonnaise) - Toothpaste Dγ - Vaseline a - b - shear thickening In addition to these shear rate- c - shear thinning dependent effects, shear time Fig. 2 Viskosity curves of fluids dependent flow behavior is ob- served in some non-Newtonian substances.

Shear thinning (pseudoplasticity) τ = f (γ,t) (1.9) At large shear rates, η decreases with the shear rate (Fig. 2, Curve c). At small shear rates, substances usually have Newtonian behavior.

11 Visco handbook The shear viscosity is thus in- Rheopexy fluenced by the duration of the Shear viscosity increases at con- shear (see Fig. 3). stant shear rate with increasing η shear time. b Rheopexy is observed, for exam- ple, in PVC plastisols. They are a used for corrosion protection of metals. Rheopexic liquids are characterized by gradual struc- c ture formation under shear.

As with the thixotropy, rheopexy ts is also only present when the a = time-independent viscosity viscosity at rest is achieved after b = Rheopexy the shearing strain after a certain c = Thixotropy time.

Fig. 3 Viscosity curves of fluids Viscoelasticity The combination of viscous and It is divided into: elastic behavior leads to the des- ignation of viscoelastic fluids. Thixotropiy In particular, polymer melts and Shear viscosity decreases at con- solutions show such properties, stant shear rate with increasing depending on the molecular shear time. At rest, the internal structure. The description of structure of the sample builds viscoelastic behavior is a central up again, so that the origi- component of the rheology of nal initial viscosity is achieved non-Newtonian liquids [1], but again after a certain rest time. without significance for visco- Many paints or varnishes show metry with glass capillary visco- thixotropic flow behavior in or- meters. der to achieve optimal flow and leveling/sagging behavior.

12 1.3 Principles of viscosity measurement Rheological measurement pro- The most important possibilities cedures are mainly using me- for the realization of the defor- chanical methods, since they are mation of the sample is shown in based on the mechanical quanti- Fig. 4. ties stress and strain of the sam- ple. Perfection in manufacturing and sophisticated quality-assurance Viscosity measurement instru- methods form the basis of stan- ments usually generate a de- dardized measurement systems fined shear deformation (shear) which meet today's highest ac- and measure the required shea curacy requirements as to repro- stress, or vice versa. The ratio of duction uncertainties and abso- the two variables is, according to lute measurement uncertainty. Eq. 1.3 the viscosity. 4 2

M1 2 3 5

1 6 2

M2 v v

a b c

a = Capillary b = Rotational viscometer c = Falling ball viscometer 1 = Capillary 5 = Measuring ball 2 = Sample 6 = Glass cylinder 3 = Coaxial cylinder M1, M2 = Timing marks 4 = transducer Fig. 4 Measurement principles of viscometers 13 Visco handbook CHAPTER 2 BASICSFUNDAMENTALS OF CAPILLARY OF VISCOMETRYCAPILLARY VISCOMETRY 2.1 Measurement principle Inside the capillary viscometers, • incompressibility of the liquid the velocity gradient required • wall adherence of the liquid for viscosity measurement is • neglect of the flow influences built up in the form of a laminar at the entry and exit of capil- tube flow within a measurement lary of sufficient length capillary. the liquid moves in coaxial lay- Under idealized conditions, ers toward the pressure drop through the capillary, in which • laminar, isothermal, steady a parabolic velocity profile is flow condition formed (see Fig. 5). • Newtonian flow behavior of the liquid • pressure independence of the viscosity

v R r max v

v = 0

Fig. 5 Velocity profile with laminar tube flow 14 The Hagen-Poiseuille Law is the whose measurement accuracy basis* for the description of the depends on the achievable ac- viscosity of all viscometers oper- curacy in differential pressure ating according to the capillary measurement as well as by the principle [3, 4]: stabilization of a defined volume 4 flow. V π R ∆ρ = (2.1) t 8Lη Fields of application are, for ex- ample, relative measurements With regard to viscosity measure- in polymer analysis, in which the ment, this results in two different pure solvent is used as a refer- fundamental measurement prin- ence liquid in a first capillary. The ciples: sample is loaded, as in a chro- matography system, in a sample 1) Measurement of the volume loop between the first and a flow through the capillary at a second capillary, which is con- given differential pressure nected in series with the first 2) Measurement of the differen- capillary - through this, the sample tial pressure at a constant volume only flows through the second flow of the liquid through the capillary and the volume flow capillary. is equal in both capillaries. The The first measurement principle pressure drop resulting from the is used in capillary viscometry volume flow is measured across with glass capillary viscometers: both capillaries and evaluated as The differential pressure is estab- a relative viscosity. lished as hydrostatic pressure in a simple manner and with very Another field of application of good reproducibility. Details re- the first measurement princi- garding this measurement prin- ple is the measurement of the ciple can be found in the next viscosity of polymer melts. The section. sample is pressed through the capillary under high pressure. Continuously operating viscom- Short capillaries are used here, eters can be built with the sec- frequently gaps with special ge- ond measurement principle, ometry (high-pressure capillary viscometry). * For real tube flows in capillary viscometers, corrections may need to be performed, which are described in Appendix A 15 Visco handbook 2.2 Designs of glass With both viscometers the capillary viscometers liquid being examined is filled through the filling tube (3) into In low-pressure capillary viscom- the reservoir (4). Since the mean eters, the viscosity is measured pressure level in OSTWALD via the flow time of a defined vol- viscometers depends on the ume of liquid through a measur- filling height, the prescribed ing capillary. measurement volumes must be strictly observed. A is The driving force is the hydrostat- therefore used for filling. The ic pressure of the liquid column. sample is sucked into the It can also be operated with over- tube (2) for measurement. It pressure to achieve higher shear measures the time required for rates. the meniscus to sink from timing

mark M1 to measurement mark M2 Irrespective of the specific de- (annular measurement marks). sign, the mostly U-shaped glass bodies have ball-shaped exten- With UBBELOHDE viscometers, sions, the volume of which de- the transition from the capillary termines the quantity of the sam- (7) into the leveling bulb (6) is ple. Measurement marks on the designed as a spherical cap. An glass body, or accurately defined additional venting tube (1) is fixed sensors, allow the mea- connected to the leveling bulb surement of the passage time of (DIN 51562-1, ISO 3105 [2, 32, 33]). the boundary layer between the After the sample is filled via the sample and the air (meniscus), a filling tube (3) in the reservoir (4), process which enables the pas- the venting tube (1) is closed. sage time of a sample volume Depending on the operational defined in such a manner to be mode, i.e. pressing or sucking, measured with measurement un- the sample is filled by over- certainties < 1/10 s. pressure applied to tube (3) or by suction via the tube (2) into Fig. 6 shows the two fundamental the capillaries (7), the measuring types of viscometers according to sphere (8), and at least up to half OSTWALD and UBBELOHDE. of the pre-run sphere (9).

16 After venting the tube (1), the During the measurement, the liquid column in the leveling liquid flowing out from the cap- bulb (6) breaks off. At the exit of illary drains on the inner wall of the capillary, the so-called sus- the leveling bulb (6) as a film. pended level develops (see also In this way, the hydrostatic pres- Fig. 22). For this reason, only a sure of the liquid column is inde- limited amount of sample may pendent of the amount of sub- be filled in (between max .- / min. stance filled in. fill marks (5)).

2 1 3

1 Vening tube 2 3 2 Capillary tube 3 Filling tube 9 4 Reservoir M 5 Min/max fill marks 1 8 6 Level bulb 7 Capillary M 2 8 Measuring bulb M 1 9 Feeder bulb 8 M h average hydrostatic 2 hm m 7 driving head L L Capillary length

M1, M2 Timing marks 5 L 6

4 7 4 a) b)

Fig. 6 Glass capillary viscometers according to a) UBBELOHDE and b) OSTWALD 17 Visco handbook In addition, owing to the geo- metrical shaping of the leveling bulb (6), the influence of surface 1 2 tension on the measurement result is almost eliminated.

In the case of the UBBELOHDE viscometer, too, the measure- ment is aimed at the time re- quired by the liquid meniscus to sink from the annular mea- 3 surement mark M1 down to the annular measurement mark M2. In the case of very strongly tinted, opaque liquids, it can be possible that a visual detection of the me- niscus passage through the mea- L surement marks is impossible M owing to the wetting of the tube. 1 In this case, reverse-flow visco- meters (see Fig. 7) are used for M manual operation (DIN 51366, 2 ISO 3105). The viscometer is filled standing over head, in which the M capillary tube (2) is immersed in 3 the sample and the sample is sucked to a fill mark (3) above the spherical bulb. The tube (1) 1 Riser tube is closed during the thermostat- 1 ic control and opened for the 2 Capillary tube start of measurement. To measure 3 Fill mark the viscosity, the flow time of the L Capillary length meniscus is captured by the mea- M1, M2, M3 Ring measurement suring marks M1, M2 and M3 on marks the riser tube (1). Additional nformation on handling can be Fig. 7 CANNON-FENSKE Reverse found in Chapter 6. Flow Viscometer 18 CHAPTER 3 3.2 Automatic timing Tasks and particularities MEASUREMENT OF In the case of automatic THE FLOW TIME capillary viscometers, an elec- 3.1 Manual timing tric signal has to be generated during the passage of the air/ In the simplest case, the flow sample or sample/air interfacial time is captured by an observer layer, respectively, through the using a stopwatch. Glass visco- measurement marks. meters manufactured for this purpose have annular measure- It is required as ment marks burnt in above and • a start and stop signal for below the measurement sphere the timing process, (see Fig. 6, 7). as well as • a status signal for automatic The disadvantages of this meth- operation (pumping up the od are: liquid into the measurement bulb, emptying of the vis- • Subjective observation errors cometer). or differences in the reaction time of the operator at the be- The detection and transforma- ginning and end of the timing tion of a time signal does not lead to increasing repeatability pose any metrological problems. uncertainties and, under certain In practical viscosity measure- circumstances, to systematic ment, the measurement uncer- errors. tainties are determined by the fluid-dynamic circumstances and • In the case of opaque substanc- the detection of the meniscus es the meniscus cannot be seen. passage through the timing One must resort to Reverse-Flow marks. viscometers with their more intricate handling and reduced accuracy.

19 Visco handbook The manufacturer of the mea- Optical sensors surement device has to ensure During the meniscus passage, by design and production pro- the optical conditions such as re- visions that the viscometer con- fraction and reflection within the stant will not change even if the detection plane are changing. measurement conditions should This leads to a change in the radi- deviate from the calibration con- ation intensity of the light arriving ditions (e.g. measurement and from the transmitter at the receiv- calibration temperature). er (see Figure 8). Detection of the meniscus For the measurement of time, passage for instance, the analogous sig- This task requires the use of sen- nal provided by a photo diode sors responding to the difference is transformed into a pulse used between the material properties for the start and stop of the time of the air and the sample being measurement. Specific threshold analyzed during the passage of values of the analogous signal the meniscus through the mea- may be defined for the "filled" or surement marks. "empty" status.

• Advantage: Versatile application; simple set- up.

• Disadvantage: Strongly colored or opaque liq- 1 2 uids, particularly with strong wall adhesion, can not be measured.

Optical sensors are housed in a measurement made of metal or plastic in viscometers 1 = Optical fibre Input from SI Analytics GmbH. The vis- 2 = Optical fibre Output cometer is fastened to a clamp Fig. 8 Arrangement of optical sensors connection in the tripod.

20 Fig. 8 shows the arrangement of • Advantage: the optical sensors in the mea- Measurement signal formation surement stand on the visco- is independent of tint, transpar- meter. The light is guided from ency and conductivity of the the stand head via fiber optic sample cables in the stand to the upper and lower measurement level. • Disadvantage: The watertight sealing enables Higher manufacturing costs the measurement stands to be (melting in of the sensors); risk placed in liquid thermostats. of encrustation and soiling with Owing to high precision in the thermally decomposable sam- glass-technological and me- ples. chanical production as well as through measures of quality as- Fig. 9 schematically shows a TC surance, it is ensured that the viscometer from SI Analytics. glass bodies and tripods are The glass-coated thermistors are freely interchangeable, with the clearly visible in the tube axis, certified viscometer constants re- whose diameter is <1 mm in maining valid. melted-in head portion.

Thermal conductivity sensors Small thermistors melted in on the level of the measurement plane (NTC resistors) are heated upper NTC - Sensor up by an electric current. Due to the better thermal conductivity of the fluid, the thermistor cools off with the air/sample transition, thereby increasing its electrical lower NTC - Sensor resistance.

Fig. 9 TC viscometer from SI-Analytics

21 Visco handbook CHAPTER 4 They can be summarized to a characteristic variable for a given METHOD FOR viscometer, the so-called viscos- ity constant K: DETERMINING VISCOSITY ν = K ⋅ t (4.2) The absolute value of the kine- Due to the unavoidable toleranc- matic viscosity is determined es when manufacturing the de- from the flow time using the cali- vices, only an approximate value bration constant. of the constant K can be calcu- lated from the dimensions of the The model of a laminar tube viscometer. The precise value of flow forms the starting point, de- K is determined by the calibra- scribed by the Hagen-Poiseuille tion of each individual viscome- Law (see Equation (2.1)). ter (see Chapter 5). The hydrostatic pressure of the According to equation (4.2), liquid column in the form of the there is a linear relationship be- mean pressure level h serves m tween kinematic viscosity and as a driving head (Images 6, 7). flow time. In Fig. 10, this is rep- Since the volume flow is detect- resented in the form of the ideal ed by measuring the flow timet characteristic curve (Curve a). the kinematic viscosity ν is ob- tained: ν

4 η π ⋅R g⋅h ν = = m t (4.1) ρ 8LV

Other than the flow time, Equa- a b tion (4.1) contains only constants ν and geometrical data. t t tg a ideal and b real viscometer characteristic curve Fig. 10 Viscometer characteristic curve 22 When using the flow model in The smaller the flow time, the the form of the Hagen-Poiseuille greater the kinetic energy cor- Law, additional pressure losses at rection time. Curve b in Fig.10 the capillary ends are not consid- shows the real characteristic ered. Due to the finite capillary curve. In the practice of viscosity length, the pressure losses in the measurement, there are basically inlet and outlet, however, have three options for taking the influ- an impact on measurement ac- ence of kinetic energy correction curacy. Due to these additional into account and thus determin- pressure losses, the measured ing the kinematic viscosity of the flow time tg is greater than the measured substance. time t, resulting from the Ha- gen-Poiseuille law. 4.1 Neglect of the kinetic

The basic hydrodynamic pro- energy correction cesses were first examined by Long flow times are obtained Hagenbach [5] and Couette [6]. by selecting viscometers with The difference between the small capillary diameter (adapt- measured and theoretical flow ed to the sample viscosity). The kinetic energy correction then time (tg- t) is therefore referred to as the kinetic energy correc- becomes so small that a correc- tion can be omitted in the con- tion time (tH) . text of the required accuracy. The minimum flow time that must t = t − t (4.3) g H be achieved in order to be able to disregard the kinetic energy This results in the following cor- correction is different, depend- rected equation for glass capillary ing on type of viscometer and viscometers: capillary size. In addition, there are also differences depending on whether the absolute value of ν = K ⋅ t − t ( g H ) (4.4) viscosity (e.g., for measurements of oils) or the relative viscosity (usually polymer analysis) is measured.

23 Visco handbook 4.1.1 Absolute measurements Short flow times below 100 s only Ubbelohde viscometer permitted for measurement with according to DIN 51562-1 automated instruments (i.e., not manually with stopwatch). The following minimum flow times are applicable for DIN Ubbelohde viscometer Ubbelohde viscometers: according to ASTM D 446 The viscometer according to Size Flow time ASTM is similar to DIN-types, 0 1000 s however, have a smaller mea- 0c 800 s surement volume. It is 4 ml for 0a 600 s capillary sizes 1 to 4B and 5 ml I 400 s for size 5. The variables used for low viscosity samples 0, 0C and Ic 200 s 0B have small measuring vol- and larger umes of 1, 2 or 3 ml.

Micro Ubbelohde viscometer The small measurement volumes according to DIN 51562-2 were chosen in order to keep the Based on a smaller measure- influence of the kinetic energy ment volume and a larger cap- correction low. Therefore, the illary length, the kinetic energy correction for all capillary sizes correction for Micro Ubbelohde within the specified measure- viscometers is smaller than for ment uncertainty can be ignored Ubbelohde viscometers of the if the flow time is a min. of 200 s. standard size. The Hagenbach correction can be neglected in the following minimum flow times:

Size Flow time MI 70 s MIc 40 s MII 30 s and larger

24 4.1.2 Relative measurements 4.2 Calculation of the The ISO 1628 standard applies kinetic energy correction for relative measurements in The manufacturer calculates polymer analysis. DIN or ASTM kinetic energy correction times Ubbelohde viscometers are rec- on the basis of the geometrical ommended for viscosity mea- dimensions as a function of the surements here. Generally in flow time and states them in the ISO 1628-1, the kinetic energy device descriptions. correction is omitted, as long as it is a max. of 3 % of the short- The basics of kinetic energy cor- est flow time - the blank value rection and its calculation are (solvent). The following minimum described in Appendix A. values ​​of the flow times are speci- fied: The kinetic energy correction can only be calculated approximately, Size (DIN/ min. flow time due to sample strew of the vis- ASTM) cometer. For accurate absolute 0/0 150 measurements, the calculated 0c/0C 100 value may therefore be max. of 0a/0B 100 1 % of the flow time. If it is higher, viscometers with smaller capillary I/1 75 size have to be used - to extend Ic/1C 50 the flow time - or the individual and larger sample strew is determined for the viscometer. For automated devices, an un- dershooting of these values is also permitted.

25 Visco handbook 4.3 Experimental deter- Implementation of correction mination of individual procedure kinetic energy correction 1) Determination of individual (according to DIN 51562-3) values ​​for the kinetic energy cor- rection with the standard liquids: The kinetic energy correction increases with decreasing flow ν times and strongly affects the t = t − i i = 1;2 (4.5) Hi gi { } measurement result. The viscom- K eter characteristic curve is also influenced by after-flow effects 2) Determination of the kinetic of the liquid and by the beginning energy correction tH for the flow disturbance of the suspended time tg through linear interpola- level for UBBELOHDE viscome- tion between the values tH1 and ters. tH2 :

If an undershooting of the mea-  1 1  surement range recommended tH = tH − K12  −  (4.6) 1   tg tg in the user instructions is un-  1  avoidable, an kinetic energy cor- rection for the viscometer must K12 thus describes the slope of be determined experimentally. the kinetic energy correction To do this, two standard liquids between the reciprocals of the of known viscosity are to be measured slow times tg1 and tg2: used, whereby the viscosity of the sample to be tested must lie t − t between the viscosity values of H1 H 2 (4.7) K12 = the standard liquids. The smaller 1 − 1 t g t g the difference of the viscosities, 2 1 the more accurate the correction method. The correction method is graphi- cally represented in Fig. 11.

26 t H c

t H2 1 t H

t H1 1 a

b

1/ t 1/t 1/t g g g 1/t g 1 2 a Calculation of the kinetic energy correction as performed in Appendix A (see equation A 1.13) b real curve of the individual kinetic energy correction c Interpolation line

Fig. 11 Individual kinetic energy correction (DIN 51562-3)

27 Visco handbook 4.4 Examples of viscosity Calculation of viscosity: determination ν = K · tg 2 2 Viscosity measurement of n-dec- ν = 0.0009801 mm / s 2 · 1234.57 ane at = 23 °C (ν ≈ 1.21 mm /s) 2 with UBBELOHDE viscometers = 1.210 mm / s

nd 1st Case 2 Case

Selection of a DIN Ubbelohde Selection of a viscometer with viscometer with the capillary 0 capillary l

2 2 K = 0.0009801 mm2 / s2 K = 0.01050 mm / s L = 90 mm L = 90 mm V = 5.7 ml V = 5.7 ml hm = 130 mm hm = 130 mm D = 0.36 mm D = 0.63 mm

Measurement range according Measurement range according user instructions: user instructions: 1.2 ... 10 mm2 / s 0.2 ...1.2 mm2 / s mean measured flow time: mean measured flow time:

tg = 116.05 s t = 1234.57 s g Calculated kinetic energy correc- The kinetic energy correction tion time as shown in Appendix time is approx. 0.3 s. This corre- A (see equation (A 1.13)) sponds to approximately 0.025% t = 0.69 s of the flow time. Therefore no H significant change is recorded Calculation of viscosity: in the measurement result by neglecting the kinetic energy ν = K · (tg - tH) correction time. ν = 0.0105 mm2 / s2 · (116.05 - 0.69) s = 1.211 mm2 / s 28 3rd Case CHAPTER 5 Selection of a viscometer with CALIBRATION capillary Ic The viscometer constant K is K = 0.03032 mm2 / s2 determined by individually cal- L = 90 mm ibrating each glass capillary vis- V = 5.7 ml cometer. SI Analytics guarantees reproducible calibration of the hm = 130 mm D = 0.84 mm highest precision through the use of reference measurement Measurement range according standards with very low measure- user instructions: ment uncertainty.

3 ... 30 mm2 / s Measurement principle SI Analytics determines con- mean measured flow time: stants by simultaneous flow mea- surement in the viscometers to tg = 39.95 s be calibrated (test specimens) and in reference measurement Calculated kinetic energy correc- standards, whose constants were tion time (according to equation determined by testing at the A 1.13) Physikalisch-Technischen Bunde- sanstalt (Physical Technical Fed- tH = 1.03 s eral Institution, PTB) in Brunswick.

The measurement range of This calibration procedure is re- the viscometer was under- ferred to as calibration by direct shot. In this case, a viscometer comparison, in contrast to cali- with a smaller capillary diame- bration with calibration liquids. ter should be used. If this is not possible, the individual kinetic An essential advantage is that a energy correction time must thus deviation of the absolute tem- be determined experimentally perature does not lead to an for precision measurements. increase of measurement uncer- tainty, as long as the temperature

29 Visco handbook in the thermostat bath is tem- Test liquids do not need to porally and spatially constant. be reference standards when Nevertheless, at SI Analytics, the calibration is performed by direct calibration temperature is set to comparison - the viscosity of the 23.00 °C ± 0.02 °C checked by a test fluids is actually determined certified . in the reference viscometer.

Realization However, Newtonian standard samples of the Deutscher Ka- The flow time of a test liquid is librierdienst (German Calibra- measured by several glass cap- tion Service, DKD) are used at illary viscometers in a thermostat SI Analytics as test liquids. The bath with a temperature uncer- measured viscosity of the test tainty of max ± 0.02 °C. fluid can be checked for its plau- sibility in this manner. Using these viscometers, there is a reference measurement stan- The constants of the device be- dard, from whose flow time the ing tested are determined from kinematic viscosity of the test flu- the kinematic viscosity of the fluid id is calculated. and the flow time (see Fig. 12).

Test specimen Reference viscometer

P1 P2 P3 R

ν = K R ( t g R - t H R )

ν K P = i tg t Pi - HPi

Fig. 12 Performing the calibration of glass capillary viscometers

30 Each calibration can guarantee Calibration of the viscometers the metrological accuracy of the at national metrological insti- viscometer constants for only a tutes limited period. We therefore rec- The calibration of viscometers ommend that the constants are is traced back to the viscosity tested on a regular basis or to of water. The following values have it checked by the manufac- are valid internationally (accord- turer. ing to ISO / TR 3666:1998): The check can be performed = 1.0016 mPas (20 °C) either by comparative measure- η ments with reference measure- = 1.0034 mm2 / s. ment standards (see above) or ν with reference standard liquids At metrological institutes, mas- from a certifiedd institute (e.g. ter viscometers are used with a DKD in Germany). However, special capillary length of 400 if regular oils are being used, mm. By comparing the flow time the limitation of the accuracy of water with those of higher vis- of the test procedure caused cosity calibration liquids in the by the uncertainty of the reg- identical viscometer as well as ular-oil viscosity indication the flow times of the same liquids should be noted. in viscometers of different capil- lary sizes (step-up procedure), See also DIN 51562-4: Viscometer the viscosity value of water is calibration and determination of ultimately used as a reference measurement uncertainty. standard for larger viscometer values in which water can not be measured.

Viscometers for industrial use usually have shorter capillaries (70-250 mm).

31 Visco handbook CHAPTER 6 Exception: Fewer bubbles occur in viscometers without a ven- HANDLING OF CAPIL- tilation tube (OSTWALD and LARY VISCOMETERS CANNON-FENSKE). These vis- cometers should be used in mea- 6.1 General guidelines surements of slightly foaming or bubbling liquids, since foam or for the selection of the bubbles impair the function of measuring system the photocells. Selection of the type of visco- meter Due to the venting tube of the UBBELOHDE viscometer, this The following viscometers from should not be used for samples SI Analytics can be used for the that undergo chemical reactions measurement of transparent liq- with the ambient air. OSTWALD uids: and CANNON-FENSKE routine viscometers are preferable here. • UBBELOHDE viscometer • OSTWALD viscometer If only a little substance or sol- • CANNON-FENSKE routine vent is available, the Micro viscometer UBBELOHDE or Micro OSTWALD viscometers can be used. These devices can therefor be used for manual or auto- High and low temperature mea- matic measurements with op- surements should always be per- toelectronic detection of the formed in UBBELOHDE visco- meniscus passage. In addi- meters for reasons of tempera- tion, TC-UBBELOHDE viscome- ture-induced volume changes of ters may be used with therm- the measurement substance. istor sensors. UBBELOHDE viscometers are to be preferred in the majority of applications over other types because of the benefits mentioned in chapter 2.

32 CANNON-FENSKE reverse flow According to DIN 51562 / ISO 3105 viscometers are available for and ASTM D446, the aspired manual measurement of opaque minimum flow time for absolute liquids. TC UBBELOHDE visco- measurements for most visco- meters are suitable for auto- meters should be 200 s. matic viscosity determination of opaque oils and emulsions. For relative measurements, flow Since the thermistor sensors are times of at least 50 seconds are glass-coated and sealed hermet- permissible according to ISO ically in the viscometer, conduc- 1628-1 for the viscometer with tive and highly aggressive liquids larger capillary number. See the can also be measured. specifications in Section 4.1.

Selection of the capillary For Micro UBBELOHDE visco- The capillary diameter (0.25 ... 10 meters, the flow time can be mm) determines the measure- shortened up to 35 s (viscometer ment range of the viscometer. no. MI) or 30 s (viscometer no. Each capillary diameter is as- M Ic and larger). For use of the signed a viscometry size number individual kinetic energy correc- (Table 1). tion as well as automatic flow time measurement, flow times to To select a viscometer, the viscos- approx. 10 s are possible using ity of the substance to be tested Micro UBBELOHDE viscometers must be estimated in which the according to some research re- flow time is roughly estimated sults [34, 35]. Such short flow (without kinetic energy correc- times should not be applicable tion) according to equation (4.2). in normal industrial laboratory practice.

33 Visco handbook

Table 1 shows an example of the measuring ranges, depending on the capillary diameter for DIN-UBBELOHDE viscometers.

Size No. Inside diameter of K (nominal) kinematic viscosity capillary [mm] [mm2/s2] range [mm 2 /s] 0 0.36 0.001 0.2 ... 1.2 0c 0.46 0.003 0.5 ... 3 0a 0.53 0.005 0.8 ... 5 I 0.63 0.01 1.2 ... 10 Ic 0.84 0.03 3 ... 30 Ia 0.95 0.05 5 ... 50 II 1.13 0.1 10 ... 100 IIc 1.50 0.3 30 ... 300 IIa 1.69 0.5 50 ... 500 III 2.01 1 100 ... 1000 IIIc 2.65 3 300 ... 3000 IIIa 3.00 5 500 ... 5000 IV 3.60 10 1000 ... 10000 IVc 4.70 30 3000 ... 30000 IVa 5.34 50 6000 ... 30000 V 6.40 100 > 10000

Table 1 Measurement range of DIN-UBBELOHDE viscometers [15]

34 6.2. Cleaning of capillary Especially when measuring ma- viscometers terials with high surface tension (e.g. aqueous media), there is Careful cleaning of the viscom- the formation of liquid droplets eter is a basic requirement for during the start-up process for exact and reproducible results. insufficiently cleaned viscome- Practice has shown that in the ters, which adhere to the wall and majority of cases, contaminants falsify the measurement result. are the cause of increased scat- That is why it is advisable to mea- tering of the flow time. The small- sure only substances with similar est quantities of microscopic properties in a viscometer. If this particles of dirt here can induce is not feasible, an especially thor- standard deviations up to several ough cleaning must be carried percent in the viscometer. out.

Particles which adhere firmly It is recommended to filter all to the capillary wall and are fre- cleaning agents before use. quently almost invisible are often Glass frits are suitable for this and the cause of systematic measure- syringe filters for nonhazardous ment errors. Such errors, which liquids. Paper filters tend to de- lead to an extension of the flow tachment of fibers and are there- time, are difficult to see from the fore not recommended. individual values of​​ a measure- ment series. The larger the cho- Initial or intensive cleaning sen capillary diameter, the lower the risk of contamination. There may be contaminants from transport or storage. In addition, In addition to solid particles, oil with a calibrated viscometer or grease films adhering to the - despite cleaning at the manu- inner wall of the viscometer have facturer - residual calibration oil an influence on the flow time. may be present. A thorough initial cleaning is therefore rec- ommended.

35 Visco handbook Suitable cleaning fluids have In many cases, it is sufficient to been proven: use concentrated sulfuric acid without the addition of chromate. • Initial cleaning This is also dangerous and it is For the removal of possibly oil essential to observe the guide- existing residues of calibration, lines for hazardous materials. calibrated viscometers should be rinsed with a volatile petroleum A milder cleaning agent is a solu- spirit (e.g., boiling range 40 °C - tion of 15% hydrochloric acid 60 °C). and 15% hydrogen peroxide. This mixture is oxidizing and can thereby remove certain organic • Intensive cleaning and inorganic contaminants. The classic cleaning agent for stubborn impurities is concen- In many cases, a laboratory trated sulfuric acid with the ad- detergent such as Mucasol® dition of potassium dichromate is successful. Such laboratory (chromic acid). cleaners are alkaline and there- fore also attack glass at high ATTENTION! concentration, at high tempera- In many , the use of tures and for long duration of chromic acid is forbidden! Ex- action. So that the calibration treme caution is always advised constant of the viscometer is not for: Chromium (VI) compounds changed, laboratory cleaners are highly toxic and carcinogenic. should only be used Handling is therefore reserved to trained laboratory personnel in • occasionally, compliance with the applicable • as a dilute solution in accor regulations. If possible, this dance with manufacturer's in- chemical should be avoided. structions, • not at an elevated temperature • and with an exposure time of max. 60 minutes.

36 Method of cleaning The use of strong alkaline sol- 1. Complete filling the viscome- vents leads to glass corrosion, by ter with the cleaning substances which the viscometer constant is specified above. changed. 2. Performance at room tem- perature. As exposure time of Daily cleaning approx. one minute is sufficient The viscometer should be in the case of petroleum spirit. cleaned with suitable solvents For intensive cleaning with sul- immediately after each measure- furic acid, chromic acid or hydro- ment. The use of a vacuum pump chloric acid / hydrogen peroxide, has been proven useful here. it is recommended that the reagent fill the viscometer at Cleaning by using a vacuum least 12 hours. pump 3. Rinse the viscometer with dis- 1. Connect the vacuum pump to tilled water. This does not apply if the capillary tube via a liquid trap. oils or similar hydrophobic sam- 2. Pour in the cleaning liquid into ple residues are removed with the filling tube and the venting petroleum spirit or similar. tube (for UBBELOHDE viscome- 4. Rinse with a filtered, water- ters). miscible, highly volatile sol- 3. Periodically close the filling vent (e.g., acetone). This is not and venting tubes during aspi- nescessary if oil or similar hydro- ration of the liquid. A pulsating phobic sample residuals or flow of liquid is created, which similar are removed with volatile dissolves even stubborn contam- petroleum spirit. inants. 5. For drying with a flow of air, 4. Repeat the cleaning process it is preferably generated by if necessary (once or twice). connecting a vacuum pump, not 5. Rinse with highly volatile sol- by compressed air: This proce- vent. dure is safer when the air is clean 6. Dry by sucking dry, dust-free and oil-free. air.

37 Visco handbook Cleaning without using a 6.3 Preparation of the vacuum pump measurement 1. Pour cleaning fluid into the fill- Sample preparation ing tube. 2. Suck the liquid several times into Solid particles in the sample to the measurement bulb. be tested have similar effects on 3. Clean the remaining viscome- the measurement result as do im- ter parts by shaking the viscom- purities in the viscometer. For this eter. reason, the following applies: 4. Drain the viscometer. 5. Repeat the cleaning process 1. All parts coming into contact two or three times. with the measured substance 6. Rinse with filtered, highly vola- should be thoroughly cleaned tile solvents. and dried. 7. Dry by blowing dry, dust-free 2. The samples have to be fil- air or in a drying cabinet. tered.

The cleaned viscometers should Filtration of low-viscosity be kept free of dust. A cleaning samples operation should also be per- • Glass filter porosity 2 to 4 (10 - formed if the measured values ​​ 100 µm). (flow times) scatter by more than • Syringe filters with porosity 0.2%. 5 µm: Such attachment filters may only be used for non- To reduce the probability of oc- hazardous samples, since there currence of such errors from the is a risk that the attachment outset, cleaning of the viscome- filter will come off during filtration ter on a regular basis with inten- and the sample thereby splashes sive cleaning fluid at longer inter- around. Syringes with luer lock vals is recommended. connections provide increased safety. • For hazardous chemicals, particularly in polyme analysis: SI Analytics therefore offers a filter device ProClean II.

38 Filtration of highly viscous Only one mark is present for samples Micro UBBELOHDE viscom- Sieve, mesh width 0.3 mm. eters, for which a tolerance Paraffin or resinous products and range of about ± 1 mm should substances in which the pour be observed. Accurate meter- point is less than 30 °C below ing is therefore not necessary. It the testing temperature should should only be ensured that the be heated accordingly before opening of the venting tube the measurement. The measure- on the reference level vessel is ment temperature must be at above the liquid level. least 20 °C higher than the pour point. Air bubbles can lead to in- creased measurement devia- Filling of UBBELOHDE and tions during the measurement process. Therefore, make sure OSTWALD viscometers that the viscometer is filled with- The substance to be tested is out bubbles. filled via the filling tube into the reservoir. To do this, the viscometer is held somewhat obliquely and the Considering that the driving liquid so poured in so that it head of the OSTWALD Visco- flows down without bubbles on meter depends on the filling the filling tube into the reservoir. quantity, the sample volumes for OSTWALD and Micro Viscometer type sample OSTWALD Viscometers indi- amount cated in table 2 are to be [ml] adhered to in any case. A pipette OSTWALD 3 is therefore used for filling. Micro OSTWALD 2 UBBELOHDE 15 - 20 UBBELOHDE viscometers have Micro UBBELOHDE 3 - 4 two fill marks on the storage CANNON-FENSKE routine ca. 7* reservoir that designate the CANNON-FENSKE opaque ca. 12* maximum and minimum filling quantity. * Filling upside-down until filling mark

Table 2 Filling quantities of various viscometers 39 Visco handbook Especially when filling in sub- stances of a higher viscosity 1 into OSTWALD viscometers, the 2 pipette should be immersed deeply into the filling tube in order to prevent errors due to flow-tailing. 6

5 Filling CANNON-FENSKE routine viscometers 8 CANNON-FENSKE Routine Vis- cometers (see Fig. 13) are held 4 upside down for filling. The capillary tube (1) immerses into the liquid to be measured, while 7 suction is applied at the other tube (2) until the liquid has reached the timing mark M2. After filling, the viscometer is 3 placed in the normal measure- ment position. 9 Since the filling process of re- 1 Capillary tube verse flow viscometers is some- what more complex, a reference 2 Vent tube should be made at this point 3 Reservoir to the standards DIN 51366, 4 Lower annular timing mark M2 ISO 3105, ASTM D446 as well as 5 Upper annular timing mark M to the user instructions. 1 6 Sphere 7 Capillary 8 Measuring bulb 9 Tube extension

Fig. 13 CANNON-FENSKE Routine viscometer 40 Suspending the viscometer in 6.4 Performing the racks measurement SI Analytics offers brackets or Thermostatization holders for all types of viscome- ters, which ensure a stable verti- Viscosity depends strongly on cal position of the viscometer in the temperature. Therefore, the the thermostat bath. They also viscometers have to be treated protect the viscometers from in a thermostat during the mea- breakage. Before measuring, surement. The thermostats used UBBELOHDE viscometers should are automatically controlled glass be inserted in the brackets pro- panelled viscothermostats. The vided for this purpose (see Fig. viscometer has to be immersed 14), and fixed in position by until the bath liquid is at least pressing the spring downwards. 2 cm higher than the liquid meniscus in the viscometer in its highest position.

The test temperature should be kept temporally and spatially constant to ± 0.02 K in the range between + 15 °C to + 100 °C. Greater deviations cannot always be avoided outside the speci- fied temperature range, but this should still not exceed ± 0.05 K. Calibrated precision thermome- ters with a resolution of 0.01 °C are recommended for tempera- ture control. Such a high degree of accuracy is often achieved by certified mercury-glass ther- mometers with a scale division of 1/100 °C.

Fig. 13 CANNON-FENSKE Routine Fig. 14 Using the viscometer in viscometer a bracket 41 Visco handbook The alternatively available elec- When using transparent thermo- tronic platinum resistance ther- stats of the CT series, special in- mometers with comparable serts for viscometer brackets are precision are considerably more available for manual measure- expensive. In addition, the valid- ment. ity of the calibration over a long period is difficult to realize. The Subsequently, the sample is ex- liquid bath and especially the posed to thermostat treatment thermometer must be protected in the viscometer. For mea- from direct light. Recommended surements with UBBELOHDE, bath fluids are: OSTWALD or CANNON-FENSKE routine viscometers, it is recom- below 0 °C alcohol mended to pump the liquids 0...80 °C distilled water at least three times into the + tap water measurement bulb in order to speed up the heat transfer. 80...105 °C water + glycol, silicone oil This is not possible with reverse flow viscometers. Their tem- 105...200 °C silicone oil, also of perature adjustment should limited suitability: polyglycols, therefore be correspondingly paraffin oil longer. The following tem- perature-adjustment times are The transparent thermostats recommended: of the CT series developed by SI Analytics for capillary viscom- 5 min moving, low-viscosity etry meet the requirements of substances viscometry in terms of temporal 10 moving, highly viscous and spatial constancy of bath min substances, low-viscosity liquid temperature of ± 0.02 °C. substances in the reverse They have openings or inserts flow viscometers for two (CT 72/2) or four capillary 15 highly viscous substances in viscometers (CT 72/4). The capil- min the reverse flow viscometers lary viscometers filled and insert- ed into the bracket or holder are The greater the measured tem- suspended in the pre-tempered perature differs from the ambient thermostat bath. temperature, the longer the tem- perature adjustment time has to 42 be chosen. Manual measurement When measuring highly viscous samples, it is recommended that To measure the flow times, the the capillary tube is kept closed liquid is sucked into the mea- after releasing the venting tube surement bulb by applying a until the leveling bulb has run vacuum to the capillary tube. empty and the suspended level When using viscometers with a has built up. feeder bulb, the latter should be filled at least up to its half. When studying volatile substanc- es, it is advantageous to realize Viscometers without a feeder the filling of the measurement bulb are filled until the liquid sphere by overpressure at the meniscus is approx. 20 mm filling tube in order to prevent above the upper timing mark. evaporation. Closing and open- If UBBELOHDE viscometers are ing the venting tube in the case used, the venting tube must be of UBBELOHDE viscometers closed prior to aspiration (e.g. should be done analogously. with finger). After completion of the filling process, the suction The measurement involves the hose is removed from the capillary period of time over which the tube and, for the UBBELOHDE lower for vertex of the meniscus viscometer, the venting tube sinks from the upper edge of the released. upper annular mark down to the upper edge of the lower annular a) b) c) mark. The stop watch used for tim- ing should have a resolution of at least 0.01 s.

When the meniscus passage is detected,it has to be made sure that the annual mark is at eye level (see Fig. 15). Perspective: (a) – correct (b), (c) – not correct, reading error due to parallax Fig. 15 Detection of the meniscus passage with manual measurement 43 Visco handbook In order to make the mea- If the flow times of a series of surement values available for measurements is scattered by statistical evaluation, the mea- more than 0.2-0.4%, one of the surement process should be re- following causes of error is usu- peated several times. Especially ally present: in the case of UBBELOHDE vis- cometers, in order to avoid any • The sample is not homoge- formation of bubbles, it should neous, or there are particles in be noted that a renewed sucking the solution. Samples with parti- or pressing up of the measure- cles can only be measured after ment substance must only begin filtration. when the drainage of the liquid • The viscometer is contaminated. from the capillary is completed. • The sample is not sufficiently thermostatically controlled. When using reverse flow vis- cometers, sucking the liquid After cleaning the viscometer, into the measurement sphere the measurements must be is not applicable. To perform repeated with a new portion of the measurement, the tube filtered sample. If only one "out- that was closed after filling lier" is present, it can be deleted, is opened on the side of the or better yet, be replaced by measurement sphere, and the an additional measurement. time over which the liquid rises An outlier test may need to be from the lower to the upper an- performed [17]. nular mark is subsequently mea- sured. The calculation of the viscosity is based on the mean value of the The CANNON-FENSKE reverse flow times. flow viscometer is equipped with two measurement spheres one on top of the other , i.e., two measurement values are avail- able after just one liquid passage. Reverse flow viscometers must be emptied, cleaned and refilled to repeat the measurement.

44 Automatic measurement For automatic viscosity mea- The correct measurement stand surement using UBBELOHDE, AVS®/S, AVS®/SK or AVS®/S-CF CANNON-FENSKE routine and for automatic viscosity measure- MICRO OSTWALD viscometers, ment is chosen according to SI Analytics offers the automatic viscometer type and the bath viscosity measurement devices liquid of the thermostats (metal of the AVS® series. Table 3 pro- stand for non-aqueous media, vides an overview of the device PVDF stand as a corrosion-free program. option).

Instrument Meniscus Way of pumping Remarks type detection sample into measuring bulb

ViscoClock light barrier vacuum or pressure, No Cannon-Fenske manual viscometers

AVS® 370 light barrier vacuum or pressure, PC controlled NTC- sensor autonatic

AVS® 470 light barrier vacuum or pressure, Stand alone, NTC- sensor automatic with printer, without PC

AVS® Pro III light barrier vacuum, Full automated NTC- sensor automatic viscosity measuring system

Table 3 Automatic viscosity measuring instruments from SI Analytics

45 Visco handbook No measurement tripod is re- The viscosity measuring devices quired for measurement using are operated either via built-in the TC UBBELOHDE viscome- software (AVS® 470) or via a PC ter. The viscometer is clamped with the appropriate PC software into a special bracket and sus- (AVS® 370, AVS® Pro III). pended in the thermostat bath. The connection with the con- The AVS®Pro III Automatic Vis- trol unit is made using a cable cosity Sampler is a fully automat- which is plugged into a socket ic viscosity measurement sys- on the viscometer head. The tem for routine measurements. viscometers are pneumatically This device performs measure- connected to the AVS® device ments of kinematic and relative via silicone or PTFE hoses. viscosity up to calculation and documentation in an self-acting All automatically operating de- manner. Filling, discharing, and vices are microprocessor con- rinsing of the viscometers are trolled. An RS-232-C interface integrated in the automatic allows for the connection of an course of the measurement. The external printer or computer. viscosity limit of the sample is The parameterization and the about 800 mm2/s at 25 °C due start of the automatic measuring to the automatic filling. sequence occurs at the control unit. The displacement of the measuring liquid in the viscom- eter is performed by an internal pressure or suction acting micro- pump. It is so controlled that an optimal pumping pressure for the reproducible filling of the measurement system is set, de- pending on the viscosity of the sample.

46 CHAPTER 7 7.1 Correctable errors and corrections SOURCES OF ERROR Surface tension AND SPECIAL correction CORRECTIONS Surface tension causes the liquid The following information is not a which is wetting the tube wall to substitute for an individual study climb by a distance of ∆h. The of the fundamental standard DIN size of the relative error ε in terms 53012 regarding this topic. We of % can be calculated on the therefore strongly recommend basis of the following formula: acquiring more information.

2  1 1  σ σ0  Many of the influences and cor- ε =  −  − ⋅100% ghm  r1 r 2  ρ ρ 0  rections named in the following (7.1) section are negligible in general practice, since they change the hm - mean pressure height measurement uncertainty only in g - acceleration due to the range of < 0.1%. r1 - radius of the measurin bulb

r2 - radius of the spherical cap Certain factors may cause signif- of the bulb, flowing into icant errors, however. Most im- the liquid from capillary portant is the temperature con- σ - surface tension of the trol of the viscometer. In addition, measurement substance the change in the gravitational σ0 - surface tension of the acceleration at the site in relation calibration substance to the calibration location as well ρ - density of the measurement as high surface tension of the substance samples may represent sources ρ0 - density of the calibration of error that should be corrected. substance

47 Visco handbook If the relation between surface Thermal expansion of the tension and density of deviates capillaries and the measure- considerably from sample to cali- ment vessel bration substance, measuremnet During high- and low-tempera- uncertainty increases: ture measurements, the radius and the length of the capillaries, a) in the case of viscometers the volume of the measurement with a small pressure head, sphere, and the average pres- where the liquid flows sure height of the viscometer will from the upper container change owing to the large differ- into another container the ence between the measurement diameter of which is con- and the calibration temperature. siderably different from the For this reason, the viscometer one of the upper container, constant has to be corrected in e.g. CANNON-FENSKE and the case of precision measure- OSTWALD viscometer. ments. b) in all viscometers where the The corrected device constant fluid flows out freely from the according to DIN 53012: capillary.

K′ = K (1+α (ϑ −ϑ0 )) (7.2) In the case of UBBELOHDE viscometers, the correction will Viscometers from SI Analytics are in general be no more than calibrated at a temperature of 0.1 to 0.2 % and can thus be ϑ0 = 23 °C. The coeffcient of lon- neglected in most cases. This gitudinal expansion α used for has been demonstrated experi- production by DURAN®-glass) is -6 -1 mentally [12]. 3.3 · 10 K .

Even at a temperature difference of 150 °C, the change in the calibration constant according to Eq. (7.2) is less than 0.1 % and can therefore be neglected in general.

48 Thermal expansion of the Inclination error measurement substance Viscometers have to be used in In the case of UBBELOHDE the position in which they were viscometers, no correction is calibrated. If the connection line required, since the measurement between the center points of the result is largely independent of reference bulbs deviates from the substance quantity being normal position, the mean pres- filled in. If, in the case of viscome- sure head of the viscometer will ters without suspended level, the change. If, instead of the initial substance temperature should angle Φ0 the connection line deviate from the measurement compared to perpendicular is temperature during the process at an angle of Φ the corrected of filling the viscometer, a vol- viscometer constant is to be ume change of the measurement calculated according to: substance leading to a change cosφ of the viscometer constants will K′ = K (7.5) cosφ occur during the temperature 0 adaptation. The constants should The brackets or holders offered then be corrected according to by SI Analytics ensure a per- Eq. (7.3) for OSTWALD and pendicular suspension of the CANNON-FENSKE routine Visco- viscometer with a deviation < 1º. meter or according to Eq. (7.4) This corresponds to a max. for reverse-flow viscometers relative constant error of 0.02 %. (DIN 53012, ASTM D 446). This means that the inclination  4V (ρ − ρ )  ′ 2 1 error can be neglected if these K = K 1+ 2  (7.3)  π Dm hm ρ2  racks are being used.  4V (ρ − ρ )  ′ 2 1 K = K 1− 2  (7.4)  π Dm hm ρ2 

Dm - mean diameter of the liquid meniscus in the reservoir vessel

ρ1 - density of the measurement substance at filling temperature

ρ2 - density of the measurement substance at measurement temperature V - filling quantity

hm - mean pressure head

49 Visco handbook Local dependence of gravity Inaccurate adjustment and The viscometers from SI Analytics measurement of temperature are calibrated at a gravitational Errors caused by inaccurate acceleration of 9.8105 m / s2. temperature adjustment or in- sufficiencies in the temperature A correction is required if the ac- stability or temperature measure- celeration of fall at the calibration ment are frequently very large, place g0 and the acceleration of since the viscosity of most of the fall at the measurement place g liquids varies largely as a function are significantly different. Equa- of temperature. tion (7.6) is to be used to calculate the corrected device constant. According to DIN 53012, it is per- g missible to correct the viscosity K′ = K (7.6) deviation to the target tempera- g0 ture at temperature deviations The acceleration due to gravity <1 °C: on Earth depends on the geo- graphical latitude and the alti- 1  dν  Uν = − ⋅  (7.7) tude above sea level: gravita- ν  dϑ  tional acceleration decreases The temperature coefficient U with decreasing latitude and ν increasing altitude. According is determined according to the to WELMEC (Western European corresponding DIN 53017. In Cooperation in Legal Metrology) practice, however, this is more it can be calculated for different complicated than working with geographic locations using the the correct set temperature. following formula, which reflects the global gravity field [24, 37]: Fully immersed calibrated ther- mometers with a resolution of g = 9.780318 (1 + 0.0053024 sin2ϕ 0.01 K should only be used for - 0.0000058 sin2ϕ - 0.000003085· h temperature measurement. The m/s2) requirements for the use of ther- mostats are described in section ϕ: geographic latitude h: height [m] above the sea level 6.4.

50 7.2 Uncorrectable errors Disturbance of the suspend- Turbulence ing level in the case of UBBELOHDE viscometers is the basic require- ment for viscosity measurement If viscosity measurements are according to the capillary princi- performed with short flow times, ple. The transition from laminar a deformation of the suspended to turbulent tube flow occurs in level may occur. This will lead capillaries (tubes) at a Reynolds to systematic measurement number of about 2300. For errors, since the average pres- capillary viscometers, turbulence sure height of the viscometer occurs even sooner due to the will change. In addition, one influence of capillary ends (with has to reckon with an increased sharp-edged capillaries, e.g., scattering of the measurement already at 1100). values within the limit ranges between the disturbed and the For a given viscometer, the Reyn- undisturbed suspended level, olds number can be calculated and the influence of surface ten- according to the following nu- sion on the measurement result merical equation (DIN 53012): will increase. Fig. 16 shows vari- ous stages of level disturbance. V Re = 63.7⋅ (7.8) 2 Table 4 gives an overview of R ⋅K ⋅ tg the limit values of the Reynolds with the units numbers and the flow times up [V] = cm3 to which in general no distur- [R] = cm [K] = mm2/s2 bances of the suspended level will occur for UBBELOHDE vis- [tg] = s cometers (normal design). Considering that kinetic energy correction will increase with an increase in the , Size no. 0c 0a I Ic one should work with a Reynolds tg [s] 100 75 60 60 number below 200 if this is Re 500 500 300 100 possible in practice.

Table 4 Limit values of tg and Re according to DIN 51562-4 51 Visco handbook Considering further that the Start-up length limits also depend on the surface One of the preconditions for tension of the liquid and the capillary viscometry is a para- shape of the capillary outflow, bolic velocity profile. For this disturbances of this kind may reason, the flow time has to be even occur in the case of some- selected in such a manner that what longer flow times. the start-length le for the forma- tion of the profile is considerably smaller than the capillary length. According to Schiller [10], the start-up length can be calculated as follows: V le = 0.015 (7.9) π υ tg

a) b) c) d)

a) no disturbance - measurement usable b), c), d) disturbance - measurement not usable

Fig. 16 Stages of disturbance of the suspended level with UBBELOHDE viscometers (according to DIN 51562-3)

52 Error due to delayed drainage Uncertainty of timing These errors are caused by the The relative error of measure- fact that a small liquid volume ment for the time measuring ΔV is adhering to the wall of the device must be less than 0.02 % viscometer above the sinking and is to be included in the mea- liquid meniscus. ΔV will increase surement uncertainty according with the viscosity and the sinking to DIN 51562-4. The uncertainty velocity of the meniscus. The of the timing device in practice is magnitude of the error is also small compared to other uncer- influenced by the wettability of tainties. the wall, the surface tension of the liquid, and the geometry of 7.3 Trouble shooting the viscometer. Depending on the constructional shape of the table device a shortening or extension Table 5 gives a summary of some of the flow times may occur. of the major errors occurring during viscosity measurements Radiation heat using glass capillary viscometers, including their possible causes To avoid an uncontrolled heating and ways of elimination. Errors up of the liquid to be tested by which can be attributed to de- heat radiation, the liquid bath is vice defects as well as improper to be protected from direct expo- use of the automatic viscosity sure to the sun or light sources. measurement devices are not Cold lights or light sources with considered. a premounted infra-red filter should preferably be used for illumination.

53 Measuring error Cause of error Elimination of error Systematic error: Temperature of thermostatic Check and - if necessary - adjust bath temperature Flow time too short bath liquid too high. Systematic error: Dirt in the capillary Discharge and clean viscometer (cf. chapter 6.2), Flow time too long repeat measurement Temperature of thermostatic Check and - if necessary - adjust bath temperature bath liquid too low. Systematic error: Kinetic energy correction For precise measurements choose viscometer with Flow time too long not used or too small smaller capillary diameter or kinetic energy correction at short flow times has to be determined experimentally with substances with similar viscosity and surface tension compared to the sample (cf. chapter 4) Systematic error: Error due to flow-tailing, Choose viscometer with smaller capillary diameter Flow time too short Kinetic energy correction too high or kinetic energy correction has to be determined at short flow times experimentally. Systematic error: Too little sample filled Refill viscometer after discharging and cleaning Flow time too short into viscometer. (cf. chapter 6.2 / 6.3). (OSTWALD-, CANNON-FENSKE- viscosimeter) Systematic error: Too little sample filled Refill viscometer after discharging and cleaning Flow time too long into viscometer. (cf. chapter 6.2 / 6.3). (OSTWALD-, CANNON-FENSKE- viscosimeter)

Table 5 Commonly occurring errors in the use of glass capillary viscometers

54 Visco handbook Visco Measuring error Cause of error Elimination of error Systematic error: Disturbance of suspended level Viskosimeter mit geringem Flow time too short at short flow Kapillardurchmesser wählen times (UBBELOHDE viscometer) (siehe Kapitel 6.1 / 7.2) Drift of flow times Drift of bath temperature Protect bath against radiation (cf. chapter 6.4); exchange thermostat, if necessary Thermostating of sample Prolong thermostating time until flow times not finished are constant (cf. chapter 6.4) Evaporation of a highly Liquid handling in viscometer by pressure mode volatile component instead of suction mode Enhanced statistical scattering Dirt in the viscometer Discharge and clean viscometer, of flow times repeat measurement Particles in the sample Discharge and clean viscometer, repeat measurement with filtrated sample – if required, use filter with smaller pore size. (cf. chapter 6.2/ 6.3) Air bubbles For samples with negligible volatility and sufficient thermal resistance: Casting out by short-time heating Discharge and clean viscometer (cf. chapter 6.2) At new filling, pay attention to avoid air bubbles (cf. chapter 6.3)

55 Continued Table 5 Commonly occurring errors in the use of glass capillary viscometers

Measuring error Cause of error Elimination of error Very high statistical scattering of Dirt at the optical sensors Remouve measuring stand out of the thermostatic flow times – automatic measurements bath, clean sensor with fabric wetted with alcohol with light barriers, up to complete malfunction. Malfunction of light barriers due Viscometer with TC sensors instead to air bubbles, sludge or of optical detection (cf. chapter 6.1) liquid lamellas

Very high statistical scattering of Incrustation of TC sensors Transparent samples: Optical detection flow times – automatic measurements with TC sensors, Opaque samples: Manual measurement with up to complete malfunction. Reverse flow viscometers. Deterioration of sensors Change viscometer Enhanced statistical scattering Beginning disturbance of Choose viscometer with smaller capillary diameter at short flow times suspended level (cf. chapter 6.1 / 7.2) (UBBELOHDE viscometer)

Fluctuating flow times Heating /cooling times of Improve control parameters of thermostat the thermostat too long

Thermostat defect Improve control parameters of thermostat Air bubbles are created when Not enough sample UBBELOHDE viscometer: pumping liquid into measuring bulb in viscometer Replenish sample of viscometer Alternatively: Discharge and clean viscometer, refill sample and repeat measurement

Commonly occurring errors in the use of glass capillary viscometers 56

Continued Table 5 Visco handbook Visco There are alternative methods CHAPTER 8 for molecular weight determina- SPECIAL APPLICATIONS tion (GPC, melt viscosity, osmosis, light scattering on solutions, field 8.1 Testing of plastics flow fractionation), but in the Introduction quality control of plastics, solu- tion viscometry has established One of the major quality features itself as the most important stan- of synthetic materials is the mean dard method: molecular weight of the polymer molecules. The molecular weight It offers high accuracy (reproduc- characterizes the chain length ibility), reliability and robustness, of the polymer molecules which at relatively low cost of acquisi- has a decisive influence on the tion and operation. processing properties of a syn- thetic material as well as on the There are different measuring mechanical properities of the tasks for different applications: final products. The stress exerted on the plas- (polymer manufacturing) tic by the processing process may lead to changes in the poly- • Determination of the mean mer (usually degradation of the chain length or mean chains). Under certain circum- degree of polymerization of stances, the properties of the the final product (raw granules) finished part might be changed to such an extent that it is no - Characterization of the longer suitable for its intended final product purpose. The average molecular - Quality assurance weight is determined by measur- - Optimization of process ing the viscosity of a dilute poly- parameters mer solution. - Preventing the production of faulty batches

57 Visco handbook Polymer processing The relative viscosity and the • Characterization of the proper- viscosity number (VN) based on ties of the starting material (raw it, the intrinsic viscosity (IV) or granulate) also the K value according to Fikentscher is calculated for - Incoming inspection polymer applications in general - Laying out systems for polymer instead of the absolute viscosity processing of the solution. - Determining optimum process parameters The work flow for the viscometry of polymer solutions can be di- • Determination of the chemical vided into three parts: and physical properties of the finished parts 1. Sample preparation 2. Measuring the viscosity - Final inspection (flow time) of the sample and - Determination of damage to pure solvent (blank value) the polymer during processing: 3. Evaluation Optimization of process param- eters For all three parts, the applicable standards must be considered Viscometry of polymer solu- for this polymer. This is particu- larly important during the sam- tions ple preparation, but also in the Determination of chain length, evaluation. The measurement of processing properties and qual- the flow time is the least critical in ity of a plastic is carried out by many cases. viscosity measurements on solu- tions of the plastics in suitable solvents using capillary viscom- eters (solution viscometry). Learn about solvent, viscometer and used standards in Table 6.

58 Type Abbr. Solvent Capillar Operating temperature Standard

Cellulose C EWN 20 °C SNV 195598 I Cuen (CED) 20 °C DIN EN 60450 Cuen (CED) 20 °C ASTM D 4243 Cuen (CED) 25 °C ISO 5351 Cuen (CED) 25 °C AST; D 1795 Cuen (CED) 25 °C SCAN CM 15:99 Cuen (CED) 25 °C TAPPI T230-0M99 Cellulose CA Dimethyl- 0c 25 °C ASTM D817 acetate chloride/ I methanol I Micro Polyamide PA Sulfuric acid II 25 °C ISO 307 (96 %) IIc Polyamide PA Formic acid I 25 °C ISO 307 (90 %) Ic Polyamide PA m-cresol II 25 °C ISO 307 IIc Polybutylene PBT Phenol/dichloro Ic 25 °C ISO 1628-5 terephthalate benzene (50 : 50) II Polycarbonate PC Dichloromethane 0c 25 °C ISO 1628-4 I Polyethylene PE Decahydro- I 135 °C ISO 1628-3 naphthalene Ic ASTM D 1601 Polyethylene PET m-cresol II 25 °C ISO 1628-5 terephthalate IIc ASTM D 4603 IIc Micro Polyethylene PET Phenol/dichloro Ic 25 °C ISO 1628-5 terephthalate benzene (50 : 50) II ASTM D 4603 Polyethylene PET Dichloroacetic II 25 °C terephthalat acid IIc Micro Polymethyl PMMA Chloroform 0c 25 °C ISO 1628-6 methacrylate I

Polyproylene PP Decahydro- I 25 °C ISO 1628-3 naphthalene Ic Polystyrene PS Toluene I 25 °C Ic Polysulfone PSU Chloroform 0c 25 °C

Polyvinyl chloride PVC Cyclohexanone Ic 25 °C ISO 1628-2 ASTM D 1243

Styrene-acrylo- SAN Ethyl methyl 0c 25 °C nitrile copolymer ketone I Styrene-butadiene SB Toluene 0c 25 °C copolymer I

Tabelle 6 Dilute solution viscometry in polymer analytics 59 Visco handbook The solution should be always The polymer concentration is made according to the respec- usually prescribed clearly in the tive underlying standard. Very standards. Often for the overall often inconsistencies in the end solution, the specification of result are simply not based in 0.5 g polymer/100ml (= 5 g/liter ) the measurement or evaluation should be found. An analytical of the sample, but in the sample balance with a resolution and preparation. accuracy of ± 0.1 mg is necessary for weighing these small sample The standards offer a variety of amounts. information, in particular regard- ing: There are two procedures for the preparation of such a solution: • the solvents • the polymer concentration (a) The solution is prepared in • the dissolution temperature volumetric flasks with defined • and the duration of dissolution volumes, for example, in 50 ml - flask. The calculated amount The exact composition should be of polymer is weighed for this. observed for solvents. Thus, for In the example at a concentra- example, according to ISO 307 tion of 0.5 g/100 ml, in 50 ml 96 % sulfuric acid or 90 % formic total volume these are 250.0 mg. acid is used for polyamide. The Deviations from the calculated water content is thus precisely initial weight must be recorded, defined in these two cases. so that a corrected concentration can be calculated. The polymer is In some cases, multiple solvents dissolved in about ¾ of the nec- are allowed (e.g., in ISO 307 for essary amount of solvent. After polyamide or in ISO 1628-5 for dissolution, it is filled up to cal- polyester). Companies or institu- ibration mark of the volumetric tions that work together should flask with solvent (Fig.17). therefore agree on the choice of solvent in order to guarantee the best possible comparability of results.

60 It should be noted: The volume (b) The solution is manufactured of the solution is temperature in standard sample dependent. The measurement using an piston , e.g. flask must therefore be filled at TITRONIC® 500 from SI Analytics room temperature. This is partic- (see Fig. 18). ularly important when the sam- ple was heated for solution be- In this procedure, the amount of forehand - then the samples cool sample can be weighed in a cer- off to room temperature before tain range, e.g., 200 to 300 mg filling. - so an exact target value does not need to be reached. The If a magnetic stir bar was inserted sample weight is entered into for solution, it must be removed the flask burette, which then cal- before filling. In fact, it must even culates the amount of solvent be rinsed with the solvent and and meters it to the polymer. This the rinsing solution collected in method is easier to use com- the measurement flask, to avoid pared to method (a). In addition, losing adhering sample. it ensures a constant concentra- tion.

Engineering plastics often con- tain glass fibers or other addi- tives. These "impurities" must be taken into account for concen- tration calculation to obtain the pure polymer content. In order to obtain a polymer amount of 250 mg, for example, the quan- tity of sample weighed in a glass fiber reinforced plastic adds up Fig. 17 Sample preparation to:

61 Visco handbook

250 mg In practice, glass fibers most m = often occur as a foreign com- sample (%) w GF ponent. The glass fiber content 1 − 100 is determined separately by incineration in accordance with standardized methods (usually wGF (%): count of glass fibres in % ISO 3451). When preparing the solution according to method b using a When ashing, a plastic sample flask burette, this formula is con- is burned in a (usually tained in the internal software to of porcelain in the practice) and take account of the additives. then annealed at temperatures of typ. 850 °C in a muffle furnace to constant weight.

Abb. 18 Sample preparation with the piston burette TITRONIC® 500 62 2. Measuring the flow time of Therefore uncalibrated viscom- the sample and pure solvent eters can be used to measure In the viscometry of , the relative viscosity. In practice, the flow time of the pure solvent a mean value is often calculated (blank value) is also measured from three flow times, which is in addition to the flow time of then converted into the relative the sample solution in almost all viscosity. The automated (AVS®) cases. Both results are converted devices from SI Analytics auto- into a relative viscosity η : matically perform repeat mea- r surements and calculations. In η K ⋅ ρ ⋅ t t most cases, the relative viscos- η = = = r ity is converted even further η0 K ⋅ ρ0 ⋅ t0 t0 into a final result (see section 3. Evaluation). η, η0 viscosity of the sample and the solvent Blank value ρ,ρ0 density of the sample and the solvent The blank value is determined t,t0 flow time of the sample for each viscometer, thus the and the solvent transit time of the solvent. The blank value should always be the Since the flow time for sample same in a given viscometer. If the and blank value is usually mea- blank value fluctuates with re- sured in the same viscometer, peated measurements, this can the calibration constant K of the be caused by: viscometer is omitted in the cal- culation of results. • The viscometer is not clean, e.g., can still be carrying resi- In addition, the density of the dues of a previous sample mea- sample and the solvent in the surement in routine use. In this examined diluted solutions are case, the blank value should be approximately equal, so that the measured once again. relative viscosity is simply the ra- tio of the flow times of sample If there are stubborn stains, the and solvent. viscometer must undergo a suitable intensive cleaning (see chapter 6.2).

63 Visco handbook • The solvent has a slightly • When the temperature chang- different composition, e.g., when es, the blank value also changes. a new of solvent was begun. A change of the com- The blank value therefore gives position is possible, in particular experienced user information with solvent mixtures such as on whether everything is OK phenol/ortho-dichlorobenzene with the entire system: This (50/50, w/w), which is used in mainly concerns the quality of accordance with ISO 1628-5 for the solvent, the cleanliness of the polyester (PET, PBT). With this viscometer and measurement solvent, there is also a risk that temperature. The blank value the phenol crystallizes when should be measured regularly, at cooling down below room tem- least when opening a new bottle perature and, in this way, chang- of solvent. es the solvent composition. Measurement of the sample The composition may change When measuring the sample even with 96% sulfuric acid solution, the run time is usually (ISO 307 for polyamide): The measured 3x as with the blank fluid is hygroscopic, that is, it value. takes on humidity while standing in air and dilutes itself in this way. According to ISO 1628-1 , the The sulfuric acid must therefore ratio of the transit time from the always be kept tightly closed. sample to the transit time of the blank value - thus the relative viscosity - should at least 1.2. The limit of 1.2 must be complied with in order to obtain sufficient accuracy for the difference be- tween the measured flow times. The value 2 is recommended as the upper limit for the relative vis- cosity.

64 In practice, however, this value is Most important are: exceeded in some cases, when • the viscosity number VN polymers with high molecular synonymous terms: weights are measured and the - reduced viscosity ηred or I polymer concentration should - according to DIN 1342-2: not be changed because it is Staudinger function Jv specified in the corresponding • Intrinsic velocity IV standards. synonymous terms: - abbr. [η] Samples containing particulates - limiting viscosity number (e.g., glass fibers) must necessar- - according to DIN 1342-2 : ily be filtered before filling into Staudinger index Jg the viscometer. For filtering, see notes in Section 6.3. In some cases, especially for PVC (according to DIN EN ISO 1628-2) If scattering of > 0.2% occur, it is the so-called probably because particles have • K value according to entered the viscometer. In these Fikentscher is determined. cases, the measurement should be repeated with a filtered sam- The evaluation and final result ple. to be used are set out in the standards for historical reasons. 3. Evaluation Thus, as a rule, for the analysis For the evaluation of the transit of PVC, the K value according times, the relative viscosity is first to Fikentscher is calculated, the calculated with Eq. (8.2) and oth- intrinsic viscosity for PET and the er variable based upon it. They viscosity number for polyamide. are listed in Summary Table 7. Again, care should be taken when comparing the measure- ment results of different labo- ratories so that "apples are not compared with oranges". The type of evaluation is therefore to be documented.

65 Visco handbook

Quantity Name

η Dynamic viscosity

ν =η / ρ Kinematic viscosity

ηrel =η /η0 Relative viscosity

η −η /η =η −1 Specific viscosity ( 0 ) 0 r

* VN =1/ c ⋅(η −η0 ) / η 0 Reduced viscosity, viscosity number

ln(ηrel ) / c Inherent viscosity

** IV = η= lim 1/ c ⋅(η −η0 ) / η 0 Intrinsic viscosity   c→ 0

a−1+ 1+ a(2 / c + 2+ a) K value according to Fikentscher K = 0.15+ 0.3⋅ c

a= 1.5⋅logηrel

* in some standards Jv or I

** in some standards also Jg or I

Table 7 Definition of terms in solution viscometry[18]

66 Intrinsic viscosity The intrinsic viscosity is obtained The intrinsic viscosity also has a by extrapolating the reduced vis- special meaning in the scientific cosity (viscosity number) at con- field: It is linked via the so-called centration 0 ("infinite dilution"), Kuhn-Mark-Houwink relationship s. Fig. 19. with the molar mass M:

η = K ⋅Ma (8.3) For applications in quality con- [ ] trol, the intrinsic viscosity mea- In this equation, the constants K surements is determined from a and a are parameters that apply single concentration (one-point to a specific polymer, solvent, measurement) and application and a particular measurement of extrapolation formulas, e.g. temperature and enable the cal- the Billmeyer formula for PET culation of the molecular weight (ASTM D4603) or the Martin for- from the intrinsic viscosity. mula for cellulose. The workload here for the determination of an IV is similar to that of a VN.

t − t 0 ηred = t0 ⋅cPolymer

VN IV, [η]

0 g/dL 0.5 g/dL cPolymer

Fig. 19 Comparison of viscosity number and intrinsic viscosity 67 Visco handbook Since all technical polymers have The description of the plastic a distribution of different molec- analysis in this chapter is only a ular weights, a mean value is ob- summary of the main points. For tained - the viscosity average Mη details regarding the individual with the viscometric determina- methods - sample preparation, tion. The constants K and a can be measurement and evaluation, found in the literature [16]. please refer to the cited standards.

For accurate determination of 8.2 Viscosity determination the intrinsic viscosity, different concentrations of polymer sam- of oils and additives ple solutions are manufactured Mineral oils consist essentially of (so-called dilution series [36]). The a mixture of hydrocarbons. They intrinsic viscosity is obtained from are used, among other things, as the extrapolation of the viscosity a lubricant, with additives, such numbers at concentration = 0. as for the improvement of the viscosity index (VI), anti-wear, SI Analytics manufactures spe- oxidation inhibitors, etc. cial viscometers for these series. Device software is also offered in Viscosity is a decisive characteris- parallel, which automatically di- tic for the flowing and lubricating lutes and measures the solution capabilities of an oil. Lubricating in steps in these viscometers via oils form a lubricating film be- connected . tween the rubbing parts in the engine which prevents direct Due to the higher effort as com- contact of solid surfaces. The vis- pared to single-point measure- cosity of the oil thus influences ments, the measurement of both the thickness of the lubricat- dilutions series mainly limited to ing film and thus the wear as well research and development. as the energy that is lost through friction.

68 The viscosity of a mineral oil var- The determination of viscosity ies greatly with the temperature: plays a major role in the produc- tion and development of doped • At low temperatures (e.g., in oils (basic oil / additive mixtures). winter, when cold-starting the engine), it must be so low that Regular viscosity measurement the oil can be pumped to the lu- ensures adequate quality con- brication points in the engine. trol in the course of production. As regards development on the • At high temperatures (e.g., in other side, the focus is on the summer, when fully opening the examination of the viscosity- throttle; extreme loads such as temperature behavior of new driving in mountainous terrain), oil/additive mixtures. In the case oil temperatures above 100 °C of used engine oils, the determi- can occur. Sufficient lubrication nation of viscosity can be used film must still also be guaran- to determine whether the for- teed here so that the lubricating mation of the lubricating film will film does not break at the friction still be sufficient even at higher points due to low viscosity. temperatures.

The life of engine oil is limited, since in operation aging and exter- nal matter are building up on the one hand (e.g. caused by oxida- tion of the basic oil, metal abra- sion, formation of soot), and the additives are becoming lean on the other (e.g. caused by the decay of the polymers owing to shearing action, oxidation, and thermal strain) [20, 21, 22].

69 Visco handbook Viscosity Index (VI) The classification of an engine One of the frequently used char- lubricating oil in so-called SAE vis- acteristics of viscosity-tempera- cosity classes is based on dynamic ture behavior (VT behavior) of viscosity at -17.8 °C (0 °F) and kine- a lubricating oil is the viscosity matic viscosity at 98.9 °C (210 °F). index VI. The VI of an oil can be Other lubricating oils should be calculated on the basis of the vis- measured at 40 °C according to cosities at 40 °C and 100 °C by ISO 3448. using tables (DIN ISO 2909). Examples for viscometers and The magnitude of the viscosity accessories from SI Analytics de- drop occurring with increasing vices are summarized in Table 8. temperature depends on the chemical composition of the re- spective oil. A minor tempera- ture dependence of the viscosity leads to a higher viscosity index. Multi-grade, engine, and gear oils are characterized by a high VI [23].

automatic measurement manual measurement Viscometer  UBBELOHDE  UBBELOHDE  CANNON Fenske routine  CANNON Fenske routine  TC-Viscometer  CANNON Fenske opaque (for oils) (for oils) Viscosity  AVS® 370, AVS® 470  Stop watch measuring  AVS®Pro III system (up to ν ≈ 800 mm2 /s at room temperature) Accessories  Thermostat und Cooler  Thermostat und Cooler

Table 8 Measurement positions for viscosity measurements on oils and additives

70 8.3 Food testing Examples of measuring tasks The raw materials, semi-finished in the food industry and finished products to be a) Determination of the viscosity processed in the food industry of beer wort and beer [26] have different rheological prop- erties and are primarily non- Beers with a viscosity > 1.7 mPas Newtonian. They depend, e.g., are hard to filter, and this leads on temperature, water content, to a reduction of the production particle size distribution (for sus- output. On the other hand, higher pensions/emulsions), mechanical viscosity has a positive effect on processing, storage and trans- richness and foam stability. port conditions. Objectives of viscosity measure- Such non-Newtonian samples ment: are not measured with glass capillary viscometers, but with, • optimization of the mashing e.g, rotational viscometers. properties • selection of filtration strategy There are, however, liquids in • quality evaluation of malt, food production which exhibit wort and beer Newtonian flow behavior, and capillary viscometer can be used b) Determination of viscosity for their viscosity measurement. of fruit and vegetables juices Raw-pressed juices with high viscosity are difficult to clarify. Viscosity is mainly affected by the pectin percentage which, in the case of concentrated fruit juices, may rise so high in the course of production that there is a danger of jellying of the contents of the tanks. Table 8 Measurement positions for viscosity measurements on oils and additives Owing to the food-physiological importance, a complete decay of the pectin is not desired. 71 Visco handbook By way of an aimed pectino- the rise in the percentage of logical decaying process in the other than saccharide. Viscosi- course of the technological ty increases with the rise in the section of the fining and clarifi- molecular mass of the solution cation process of the juices one components (mono- and disac- tries to adjust an optimum pectin charides, glucose syrup [30]) percentage [27]. and can be approximated ac- cording to a logarithmic mixing Objectives of viscosity measure- rule: ment: η = wA logη A + w B logη B (8.4) • gathering of control parame- w - weight fraction ters for the pectinological pro- A,B - components cess of optimizing clarification and fining • quality surveillance Glucose syrups are characterized • characterization of the jelly- by different saccharide fractions, ing capabilities of pectins, inter a fact which results in diverg- alia by the determination of the ing viscous behavior patterns. limiting viscosity number [28] Considering that they are used as crystallization inhibitors in c) Viscosity determination in the production of confectionery, sugar industry viscosity is a major technological parameter. Information about viscosity is es- sential in the extraction and tech- Objectives of viscosity measure- nical processing of sacchararose ment: solutions, [29]. It increases in the form of an exponential curve with • gathering of control parameters rising concentration and has thus for processing sugar solutions a substantial influence on the • quality surveillance crystallization readiness of sugar • development of recipes solutions. So it is that the crys- • provision of information for the tallization of sacchararose solu- rating of appliances and apparatus tions is favored with increasing for sugar industry concentration (state of over-sat- uration), but will decrease with 72 d) Viscosity determination in e) Viscometry for special milk industry food applications Owing to the differences in the After examining the question of provenience and composition of knowing whether the food liquid milk and dairy products, the rhe- to be analyzed can be reason- ologic behavior of dairy products ably treated as a Newtonian fluid, differs greatly [31]. all types of capillary viscometers can be used in principle. The viscosity of milk, cream, con- densed milk etc. is influenced There may be some difficulties in by the fat contents, the concen- the detection of the liquid menis- tration of the dry matter, and, cus. Owing to their low degree of to a high degree, by the process- transparency and the after-flow ing conditions, e.g. by homo- effects, optical detection of dairy genization. An addition of hydro- products is problematic. colloids (thickening, binding, and jellying agents) and The use of TC- Viscometers re- stabilizers has a highly viscosity- quires frequent, thorough clean- raising effect. Viscosity mea- ing, since the thermistors tend surement provides valuable to soil as a result of incrustation. information required to reveal There are less problems in the their chemical structure and their viscosity measurement on beer, effect in combination with com- fruit juices and the like. Owing ponents of milk. to the fact that these fluids have a tendency of foam formation, Objectives of viscosity measure- OSTWALD Viscometers and ment: Micro OSTWALD Viscometers have proven their suitability • quality evaluation for use [26]. Similarly good ex- • development of recipes periences have been made with viscosity measurements of fruit juices.

73 Visco handbook APPENDIX A FUNDAMENTALS OF HC CORRECTION The Hagen-Poiseuille equation L (see. Chapter 2, Equation (2.1)) is l not sufficient to describe the real e flow in capillary viscometers.

Fig. A 1 shows the real pressure radient that occurs in the capillary [7]. The deviations from the ideal curve resulting from ∆p hydrodynamic processes in the entry and exit zones of the cap- illary. They are accounted for in the flow model (Fig. A 2) using additional terms. Fig. A 1 axial pressure curve in thel capillary

∆ρ

Hagen-Poiseuille equation Kinetic energy correction Couette correction (Hagenbach)

Viscous Pressure loss due to increase Pressure loss due to developing contribution of kinetic energy of liquid at a parabolic velocity profile in the the entry of the capillary entry zone le

8 η V L ρ 2 V Pressure loss see text p = 4 + + π R π 2 R 4 (A 1.1)

Fig. A 2 Flow model with correction terms

74 Hagenbach [5, 14] for the first This effect was originally included time pointed out that the kinetic by a fictitious extension of the energy of the formed capillary capillary to a certain multiple n flow - for which part of the hy- of the capillary radius n·R. The drostatic pressure must be ap- resultut is the following corrected plied - should be considered as Hagen Poiseuille equation: a correction. This energy is not ∆ pπ R 4 mρ V converted into heat the capillary η = − by viscous friction, but only after 8V (L + nR) 8π (L + nR) the outlet. (A 1.2)

This kinetic energy or the pres- The values of n lie between 0.2 sure loss can be calculated in and 1.2 - -shaped capillary principle (Fig. A2, Eq. A (1.1)). ends have lower values than Since the hydrodynamics more sharp-edged capillary ends. complicated in practice, the Since the Couette correction is kinetic energy correction term is hardly determined by measure- expanded by a constant m [2, 8]. ment, it is not calculated inprac- tice with an extension n · R of the Couette described an additional capillary. Instead the Hagenbach pressure loss [6]: A capillary with and Couette correction are sum- trained parabolic flow profile has marized in the factor m. the lowest pressure loss - any other form of flow consumes With the equations more energy. Since the parabolic ∆ p = ρ gh (A 1.3) flow profile is formed only in the m inlet zone, the pressure drop per and capillary length is greater than in V  (A 1.4) the subsequent capillary. V = tg For the kinematic viscosity can be written:

4 π R gh m mV ν = tg − 8LV 8π Ltg (A 1.5)

75 Visco handbook The parameter m is highly de- For sharp-edged capillary ends, pendent on the shape of the cap- a constant value of m = 1.12 was illary end and on the Reynolds theoretically calculated [9, 10, 11]. number (Re). This value is also specified in DIN 53012 as a maximum bench- The Reynolds number is an im- mark. Ideally, sharply cut capillary portant dimensionless para- ends are not feasible for produc- meter for fluid mechanical tion reasons. descriptions of incompressible fluids: The calculated value has been experimentally confirmed for ρ⋅v ⋅ 2 R ν ⋅2R Re = m = (A 1.6) Re > 100. For Reynolds numbers η ν below 100, m drops off sharply and has at Re = 25 only about vm = mean velocity 30 - 40% of its initial value [12]. At It characterizes the type of flow, Re <10, m is negligibly small [13]. laminar or turbulent, due to influences of inertia and friction If the capillary ends are funnel- (viscosity). Depending on the shaped, m is thus a function of production technology, the cap- the Reynolds number over the illary ends of viscometers can be entire flow time range used me- sharp or funnel-shaped (Fig. A 3). chanically. Cannon, Manning and Bell [14] found a proportion- ality between m and the root of Reynolds number: a b m = 0.037 Re (A 1.7) Equation (4.15) represents the basis for calculating the kinetic energie correction by means of a universally applicable formula for viscometers with funnel-shaped a - sharp-edged b - funnel-shaped capillary, see below. Thus, the Fig. A 3 following equations work for vis- Capillary ends of viscometers cometers with sharp-edged or funnel-shaped capillary: 76 sharp-edged capillary ends The kinetic energie correction calculated according to this B ν = K ⋅ t − (A 1.8) formula was tabulated in old t g versions of DIN 51562-1 for DIN Ubbelohde viscometers 1.12V and is included in the still valid B = (A 1.9) 8π L DIN 51562-2 of 1988 for Micro Ubbelohde viscometers. B tH = (A 1.10) K tg In the current DIN 51562-1:1998 the individual determination funnel-shaped capillary ends of kinetic energie correction is recommended. Capillary viscometers from SI Analytics have funnel-shaped The ASTM D 446 still allows the capillary ends for manufacturing use of the calculation formula, reasons but also recommends an kinetic energie correction. The reason For validity of the equation A (1.7) for this is that the kinetic energie according to [14], the kinetic correction of the individual vis- energie correction can be calcu- cometer is subject to significant lated as follows: model deviations, e.g., by the E individual shape of the inlet and ν = K ⋅ t − (A 1.11) 2 outlet funnel on the capillary tg ends. For this reason, the cal- 3 0.00166 V 2 culation formula of the kinetic E = 1 (A 1.12) energie correction should only L(2 K R) 2 be applied for small correction values, as indicated in the instruc- E tions of the viscometer. tH = 2 (A 1.13) K tg This calculation formula is used by SI Analytics by default both in the user instructions for the vis- cometer as well as in software.

77 Visco handbook SYMBOLS AND UNITS USED A Surface parallel to direction of flow m2

B Constant of the kinetic energy correction with sharp-edged capillary ends mm2 s c Concentration g/cm3

γ Shear rate 1/s

E Constant of the kinetic energy correction with funnel-shaped capillary ends mm2 s

F Force in the direction of flow N g Acceleration of fall at the place of measurement m/s2

g0 Acceleration of fall at place of determination of the viscometer constant m/s2

hm Mean hydrostatic pressure height cm

∆h Capillary rising head of the liquid cm

IV, [η] Intrinsic viscosity, limiting viscosity number cm3/g

K Viscometer device constant mm2/s2

2 2 KP Viscometer device constant (device being tested) mm /s

KR Viscometer device constant (reference viscometer) mm2/s2

K´ Corrected viscometer device constant mm2/s2

L Length of the capillaries cm

le Entry length of capillary cm

78 m Coefficient of the kinetic energy correction - n Coefficient of the Couette correction -

∆p Acting pressure mbar

∆pC Pressure loss resulting from the Couette correction mbar

R Radius of the capillaries cm

Re Reynolds number -

r1 Radius of the upper bulb on the liquid meniscus cm

r2 Radius of the lower bulb on the liquid meniscus cm s Temperature error K t Flow time corrected according to Hagenbach-Couette s

tg Measured flow time s

tgP Measured flow time (device being tested) s

tgR Measured flow time (reference viscometer) s

tH Kinetic energy correction time s

tHP Kinetic energy correction time (device being tested) s

tHR Kinetic energy correction time (reference viscometer) s

ts Shear time s

T Measurement temperature K

T0 Calibration temperature K

U Voltage V 79 Visco handbook U Temperature coefficient of kinematic viscosity 1/K ν

V Flow through volumes cm3

V. Volume flow rate cm3/s

∆V on the inner walls of the viscometer adhering liquid volume cm3

vm Average flow velocity m/s x Coordinate flow direction m y Coordinate in the direction of the velocity gradient m

α Linear expansion coefficient 1/K

ε Relative error of measurement value %

η Dynamic viscosity mPa s

ηr Relative viscosity -

η0 Dynamic viscosity of the solvent mPa s

ν Kinematic viscosity mm2/s

ρ Density of the liquid to be measured g/cm3

ρ0 Density of the normal liquid or the solvent (in the polymer analysis) g/cm3

σ Surface tension of the liquid to be measured mN/m

σ0 Surface tension of the normal liquid mN/m

ϑ Temperature °C

τ Shearing strain Pa

80 Φ Angle between the perpendicular and the line connecting the upper and lower the central point of the leveling bulbs during measurement -

Φ0 Angle between the perpendicular and the line connecting the upper and lower central point of the leveling bulbs during calibration -

81 Visco handbook BIBLIOGRAPHY

[1] Mezger, T. The Rheology Handbook, Vincentz-Verlag, 2nd ed. 2006

[2] Viswanath, D.S. et al. Viscosity of Liquids, Springer-Verlag, Dordrecht, 2007

[3] Prandtl, L. Führer durch die Strömungslehre, Springer-Verlag, Berlin, 13. Aufl. 2012

[4] Eck, B. Technische Strömungslehre Band 1, Springer-Verlag, Berlin, 9. Aufl. 1988

[5] Hagenbach, E. Poggendorffs Annalen der Physik 109 (1860), 385

[6] Couette, M. Annales de Chimie et Physique 21 (1890), 433

[7] Hengstenberg, J./ Sturm, B./ Winkler, O. Messen, Steuern und Regeln in der chemischen Technik B. II, 432 - 499, Springer-Verlag, Berlin, 3. Aufl.1980

[8] Dinsdale, A./ Moore F., Viscosity and its measurement, Chapman and Hall, London (1962)

[9] Boussinesq, V. S. Comptes rendus 110 (1890), 1160, 1238

[10] Schiller, L. Forschung auf dem Gebiete des Ingenieurwesens (1922), H. 248

[11] Riemann, W. Journal of the American Chemical Society 50 (1928), 46

[12] Weber, W./ Fritz, W. Rheologica Acta 3 (1963), 34 82 [13] Dorsay, N. E. The physical review 28 (1926), 833

[14] Cannon, M. R.; Manning, R. E. Bell, J. D. 32 (1960), 355

[15] Gebrauchsanleitung UBBELOHDE-Viskosimeter mit hängendem Kugelniveau SI-Analytics, Mainz

[16] Grulke & Grulke, Wiley Polymer Handbook, 5th ed. 2014

[17] Doerffel, K. Statistik in der analytischen Chemie, Wiley-VCH Verlag 1990

[18] Brown, R. P. Taschenbuch der Kunststofftechnik. Carl Hanser Verlag, München 1984

[19] DIN - Taschenbuch 398 Rheologie Beuth Verlag Berlin, 2 Aufl. 2012

[20] Jentsch, C. Chemie in unserer Zeit 12 (1978), 57

[21] Bannister, K.E. Lubrication for Industry, Industrial Press, Inc., 2nd ed. 2006

[22] Klein, J./ Müller, H. G. Erdöl und Kohle, Erdgas, Petrochemie 35 (1982), 187

[23] Stepina, V./ Vesely, V./ Trebicky, Vl. Tribologie und Schmierungstechnik 34 (1987), 113

[24] Gravity Zones. In: Directive 90/384/EEC - Common Application. WELMEC 2 (Issue 3),2000, S 20-24.

[25] Wilke, J./ Kryk, H. Temperaturkonstanz und Temperaturverteilung in Flüssigkeitsther- mostaten, VDI/VDE-GMA-Tagung TEMPERATUR ´92 VDI Berichte 982 (1992), 265 - 268 83 Visco handbook

[26] Greif, P. Monatsschrift für Brauerei 32 (1979), 356

[27] Köller, M./ Grandke, I. Lebensmittelindustrie 24 (1977), 162

[28] Kunzele, H./ Kleiber, U. / Bergemann, U. Lebensmittelindustrie 36 (1989), 78

[29] Hoffmann, H./ Mauch, W./ Untze, W. Zucker und Zuckerwaren, Verlag Paul Parey, Berlin 1985

[30] Schiweck, H./ Kolber, A. Gordian 72 (1972), 41

[31] Kessler, H. G. Lebensmittel- und Bioverfahrenstechnik, Molkereitechnologie Verlag A. Kessler, Freising 1988

[32] UBBELOHDE, L. Zur Viskosimetrie S. Hirzel Verlag, Stuttgart 1965

[33] UBBELOHDE, L. Öl und Kohle vereinigt mit Erdöl und Teer 12 (1936), 949

[34] Kryk, H./ Wilke J. GIT Fachzeitschrift für das Labor 5 (1994), 463

[35] Kryk, H./ Wilke J. Erdöl und Kohle, Erdgas, Petrochemie 47 (1994), 467

[36] Elias, H.-G.Makromoleküle, Band 2: Physikalische Struktur und Eigenschaften, 6. Aufl. 2000

[37] Thulin, A.: A "Standardized" Gravity Formula. OIML-Bulletin Nr. 127, 1992, S.45

[38] Gebrauchsanleitung UBBELOHDE-Viskosimeter (DIN), SI Analytics Mainz

84 STANDARDS Fundamentals

DIN 1342-1 Rheological concepts

DIN 1342-2 Newtonian liquids

DIN 53017 Determination of the temperature coefficient of viscosity of liquids

ASTM D 445 Standard Test Method for Kinematic Viscosity of Transparent and Opaque Liquids and Calculation of Dynamic Viscosity

Metrology

DIN 51366 Measurement of kinematic viscosity by means of the Cannon-Fenske viscometer for opaque liquids

DIN 51562-1 Measurement of kinematic viscosity by means of the Ubbelohde viscometer

DIN 51562-2 Determination of kinematic viscosity using the Ubbelohde microviscometer

DIN 51562-3 Determination of kinematic viscosity using the Ubbelohde viscometer, viscosity relative increment at short flow times

DIN 51562-4 Measurement of kinematic viscosity by means of the Ubbelohde viscometer, Viscometer calibration and determination of the uncertainty of measurement

85 Visco handbook DIN 53012 Capillary viscosimetry of newtonian liquids - Sources of errors and corrections

ISO 3105 Glass capillary kinematic viscometers - Specifications and operating instructions

ASTM D 446 Standard Specifications and Operating Instructions for Glass Capillary Kinematic Viscometers

BS 188 Methods for determination of the viscosity of liquids

Testing of mineral oils and used materials

DIN ISO 2909 Petroleum products - Calculation of viscosity index from kinematic viscosity

DIN ISO 3448 Industrial liquid lubricants - ISO viscosity classification

DIN 51563 Testing of Mineral Oils and Related Materials - Determination of Viscosity Temperature Relation - Slope m

DIN EN ISO 3104 Petroleum products - Transparent and opaque liquids - Determination of kinematic viscosity and calculation of dynamic viscosity

Testing of plastics and cellulose

DIN EN ISO 1628-1 Determination of the viscosity of polymers in dilute solution using capillary viscometer - Part 1: General principles

86 DIN EN ISO 1628-2 Determination of the viscosity of polymers in dilute solution using capillary viscosimeters - Part 2: Poly(vinyl chloride) resins

DIN EN ISO 1628-3 Determination of the viscosity of polymers in dilute solution using capillary viscometers - Part 3: Polyethylenes and polypropylenes

DIN EN ISO 1628-4 Determination of the viscosity of polymers in dilute solution using capillary viscometers - Part 4: Polycarbonat

DIN EN ISO 1628-5 Determination of the viscosity of polymers in dilute solution using capillary viscometers - Part 5: Thermoplastic polyester (TP) homopolymers and copolymers

ISO 1628-6 Bestimmung der Viskositätszahl und der Grenzviskositätszahl; Teil 6: Methylmethacrylatpolymere

DIN EN ISO 307 Polyamides - Determination of viscosity number

EN ISO 1157 Cellulose acetate in dilute solution - Determination of viscosity number and viscosity ratio

DIN 54270-1 Testing of textiles; determination of the limit-viscosity of celluloses, principles

DIN 54270-2 Testing of textiles; determination of the limit-viscosity of celluloses, Cuen-procedure

DIN 54270-3 Testing of textiles; determination of the limit-viscosity of celluloses,

EWNNmod(NaCl)- procedure

87 Visco handbook ISO 5351 Pulps - Determination of limiting viscosity number in cupri-ethylenediamine (CED) solution

DIN EN 60450 Measurement of the average viscometric degree of polymerization of new and aged cellulosic electrically insulating materials (IEC 60450)

SNV 195598 Textilien; Prüfung auf Faserveränderung und Faserschädigung; Bestimmung der Viskositätszahl von Cellulose in EWN-Lösungsmittel

TAPPI T230 om-99 Viscosity of pulp (capillary viscometer method)

ASTM D 2857 Standard Practice for Dilute Solution Viscosity of Polymers

ASTM D 789 Standard Test Methods for Determination of Solution Viscosities of Polyamide (PA)

ASTM D 1243 Standard Test Method for Dilute Solution Viscosity of Vinyl Chloride Polymers

ASTM D 1601 Standard Test Method for Dilute Solution Viscosity of Ethylene Polymers

ASTM D 4603 Standard Test Method for Determining Inherent Viscosity of Poly (PET)

ASTM D 4243 Standard Test Method for Measurement of Average Viscometric Degree of Polymerization of New and Aged Electrical Papers and Boards

ASTM D 4878 Determination of Viscosity of Polyols

88 Standards for liquid thermostats DIN 12876-1 Electrical laboratory devices - Laboratory circulators and baths - Part 1: Terms and classification

DIN 12876-2 Electrical laboratory devices - Laboratory circulators and baths - Part 2: Determination of ratings of heating and refrigerated circulators

DIN 12876-3 Electrical laboratory devices - Laboratory circulators and baths - Part 3: Determination of ratings of laboratory baths

89 We express our core competence with our company name SI Analytics - the manufacture of analytical instruments. In addition, SI represents the main products of our company: sensors and instruments.

Arising from the history of SCHOTT® AG, SI Analytics has nearly 80 years of experience in glass technology as well as in the development of analytical instruments. As always, our products are manufactured in Mainz with high standards of innovation and quality.

As an independent company in Mainz for over 40 years, we are, as a former subsidiary of SCHOTT® AG, still connected to tradition and, like before, continue manufacturing as a Mainz glass maker.

In the future, our electrodes, titrators and capillary viscometers will also be at home everywhere know-how is required in analytical measure- ment technology.

Since 2011, SI Analytics has belonged to the publicly traded company Xylem Inc., headquartered in Rye Brook / NY, USA. Xylem is a leading global provider of solutions for problems relating to water. We are Xylem Analytics

Xylem consists of different business sectors - such as Applied Water Systems and Analytics. The following companies make up Xylem Analytics and act like SI Analytics in the chemical, pharmaceutical, biotech, food and plastics industries.

Bellingham + Stanley • Refractometers • Polarimeters • Certified Reference Materials www.bellinghamandstanley.com ebro • Oil Quality Measurement Devices • Precision • Temperature, Pressure and Moisture Data Loggers www.ebro.com

OI Analytical • Total Organic Carbon and Cyanide • Phosphoric Organic & Organochlorine Pesticides • Volatile Organic Compounds (VOC) www.oico.com

More Xylem Analytics brands: Xylem |ˈzīləm|

1) The tissue in plants that brings water upward from the roots; 2) a leading global water technology company.

We’re a global team unified in a common purpose: creating innovative solutions to meet our world’s water needs. Developing new technologies that will improve the way water is used, conserved, and re-used in the future is central to our work. We move, treat, analyze, and return water to the environment, and we help people use water efficiently, in their homes, buildings, factories and farms. In more than 150 countries, we have strong, long-standing relationships with customers who know us for our powerful combination of leading product brands and applications expertise, backed by a legacy of innovation.

For more information on how Xylem can help you, go to www.xyleminc.com

SI Analytics GmbH

Hattenbergstr. 10 55122 Mainz Germany

Phone: +49.(0)6131.66.5111 Fax: +49.(0)6131.66.5001 E-Mail: [email protected] Internet: www.si-analytics.com

presented by

SI Analytics is a trademark of Xylem Inc. or one of its subsidiaries. © 2015 Xylem, Inc. 980 091US Version 11/2015