WO1996034269A1

WO1996034269A1 - Improved differential pressure capillary viscometer and analytical process

Info

Publication number: WO1996034269A1
Application number: PCT/US1996/004830
Authority: WO
Inventors: Scot D. Abbott
Original assignee: E.I. Du Pont De Nemours And Company
Priority date: 1995-04-28
Filing date: 1996-04-10
Publication date: 1996-10-31

Abstract

A means for measuring either the intrinsic or inherent viscosity of a solute in solution with a solvent which is independent of flow rate and temperature fluctuations. The solvent is passed through one capillary tube (10) and the solution through a second capillary (12) in series with the first, and the pressure drop across each is measured. Signals corresponding to the ratio of the pressure drop measurements and to the flow rate (as derived from the pressure drop in the first capillary) are fed to a signal processing means where they are processed to determine either the intrinsic or inherent viscosity of the solute independent of flow rate and temperature fluctuations. The viscosity measuring means may also be used together with size exclusion chromatography to determine the molecular weight distribution of polymer materials.

Description

TΠTJE

IMPROVED DIFFEREJNTIAL PRESSURE

CAPILLARY VISCOMETER AND ANALYTICAL PROCESS

FIELD OF THE INVEJNTION This invention relates to capillary viscometers. More specifically, it relates to differential pressure capillary viscometers which may be used alone to measure the viscosity of fluids or in combination with a chromatograph to obtain molecular size distribution information,

BACKGROUND OF THE INVENTION Accurate measurements of fluid viscosity are important in many industries.

A capillary viscometer is often used to measure the absolute viscosity of a given fluid. For many purposes, however, it is necessary to know the relative viscosity of two fluids.

The viscosity measuring means may also be used to measure the relative viscosity of a sample liquid and a reference liquid or to determine the viscosity of the sample liquid independent of flow rate and temperature fluctuations.

Relative viscosity is of particular importance in polymer research and manufacturing since it can be used to measure molecular weights and to determine molecular weight distributions, which provide important information relating to the physical properties of polymers. A comparison of the viscosity behavior of two polymers of the same molecular weight, for example, is used as a measure of the degree of branching. One of the oldest means used to obtain such information is to measure the viscosity of a known concentration of a polymer in a solvent. By using the ratio of the viscosity of the polymer-solvent solution, η_p, to that of the pure solvent, ^*η_s, the intrinsic viscosity [η] of the polymer can be calculated in accordance with the following mathematical relationships:

Relative Viscosity: η_r = T)p/η_s

Specific Viscosity: η^ = η_r - 1

Inherent Viscosity: i"^ = (lnη_r)/C Intrinsic Viscosity: [η] = lim i ^ =lim T /C c->0 c->0

= lim(lnη_r)/C c->0 where C = polymer weight concentration, and lim = mathematical symbol meaning the limit of the quantity when the concentration, C, approaches zero. Thus, the intrinsic viscosity [η] will approach the inherent viscosity i}^ and will also approach TJsp/C, as the polymer weight concentration C approaches zero.

Inherent and intrinsic viscosities are important polymer characterization parameters. The intrinsic viscosity, for example, provides an indication of the size of the polymer molecules. The value of [η] is not a function of polymer concentration or the viscosity of the solvent medium. The value of [η] for a linear polymer in a specific solvent is related to the polymer molecular weight M through the Mark-Houwink Equation: [η] = KMa

where K and a are Mark-Houwink viscosity constants, some of which are available in polymer handbooks.

Viscosity measuring means may also be used together with size exclusion chromatography to determine the molecular weight distribution of polymer materials.

Relative viscosity is often determined experimentally by measuring the absolute viscosity of each fluid separately with a viscometer and then calculating the ratio of the absolute viscosities. Prior art viscometers, however, have been designed to measure viscosities in a number of ways. U.S. Pat. No. 4,627,271 and U.S. 4,578,990, assigned to the same assignee as the present application, disclose a method and apparatus for measuring either intrinsic or inherent viscosity of a solute in solution with a solvent which is independent of flowrate and temperature fluctuations. In one embodiment, the solution is passed through a first capillary tube and the solvent through a second capillary tube, which are in series and subject to the same flowrate, while the pressure drop across each is measured by a pressure sensor. Signals corresponding to each pressure drop measurement are fed to an amplification means, such as a logarithmic amplifier, analog divider or high speed computer, where they are processed to determine either the mtrinsic or inherent viscosity of the solute independent of flowrate and temperature fluctuations.

T e disclosed method and apparatus can usually achieve a precision of about 1/1000. The pressure signals for the first (sample) capillary tube Si and the second (reference) capillary tube S2 can be expressed as follows:

Sι = K₁F,η₁ + Z₁

S2 = K2F2TI2 + Z2 where K = capillary constant F = flowrate η = viscosity and Z = signal value at no flow, i.e. no-flow zero value Solving for the relative viscosity yields:

τ_{Jl =} (K_?/K,HS, - Z_l) η₂ (S₂ - Z₂)

The relative viscosity η_r obtained is thus independent of flowrate, since the flowrates through each capillary are designed to be the same.

The precision of relative viscosity measurements using the method and apparatus disclosed in these patents is still subject to a number of limitations.

Problems can arise, for example, when the actual flow through the capillaries is not exactly identical. This can occur for reasons such as slight pulsations in the flow due to the pumping means, hydraulic capacitances, normal variations that would be expected to occur at a specific pump speed, as well as normal variations in upstream versus downstream flows.

Additionally, the method and apparatus disclosed in these patents are subject to limitations regarding the pressure sensors themselves, typically pressure transducers. In particular, fluctuations or errors in the no-flow zero values, Zi and Z2, for the pressure transducers can significantly affect the precision of the relative viscosity measurements, especially when small changes in relative viscosity need to be detected. The no-flow zeroes are measured for die sample and reference transducers at no flow through the capillaries. Most transducers are sensitive to absolute pressure, and since the absolute pressure measured is a function of flowrate, the no-flow zero values are also a function of flowrate. This means the no-flow zero values are only accurate within a limited flowrate range, causing error in the signals outside that flowrate range. The signals Si and S will therefore have different errors at different flowrates, which is known as "no-flow zero error". Thus, a need exists to hold the flow as close to constant as possible and eliminate error caused by variations in the no-flow zero values. Furthermore, polymeric solutions and polymeric melts can be expected to exhibit non-Newtonian fluid behavior, the extent depending on the magnitude of the viscosity, i.e., the molecular weight. When viscous polymeric samples are introduced into a system, a hydraulic upset occurs. The result is that error may be introduced into the viscosity measurements due to both transducer limitations and the rheological behavior of the fluid.

An objective of the present invention, therefore, is to provide a method and means for measuring intrinsic or inherent viscosity of a solute in solution independently of flowrate and temperature fluctuations, in order to overcome problems due to variations in actual flowrate through a sample and reference capillary, thereby improving the precision of the viscosity measurements.

Another objective of this invention is to provide continuous flow control in an apparatus for accomplishing the above process in order to improve d e precision of the viscosity measurements.

SUMMARY OF THE INVENTION The present invention provides a method for measuring the relative viscosity of a first and second fluid or liquid independently of flow rate and temperature fluctuations, which method comprises die steps of:

(a) passing the first fluid through a first capillary tube;

(b) passing the second fluid through a second capillary tube connected in series with the first capillary tube;

(c) measuring, essentially simultaneously, die pressure drop, DP_S, across die first capillary tube and the pressure drop, DP_p, across the second capillary tube when, respectively, full of flowing first fluid and flowing second fluid;

(d) generating signals corresponding to the pressure drop across each capillary tube; and (e) feeding said signals to an analogue or digital signal processor which processes the ratio of DPp DP_s for use in measuring the relative viscosity of the fluids, wherein the improvement comprises

(f J determining the flow rate of the flowing first and second fluids by processing a pressure drop signal; and

(g) transmitting a flowrate signal to a controller for limiting fluctuations in the flowrate of the fluids.

In step (e), d e signal processor may be based on real time signal processing which processes die ratio of DPp/DP_s in real time. In one embodiment of the present invention, d e first fluid is a solvent

(including a mixture of one or more individual solvents) and the second fluid is a ^ solution of a solute in said solvent. In still another embodiment of the present invention, the solute is a polymer or polymer composition, for example, having an molecular weight distribution. A measurement of the intrinsic or inherent viscosity of the solute can be obtained based on die relative viscosity. The solute may be a fluid or a solid in pure form.

The present invention also provides an apparatus for accomplishing the above process, comprising: (a) a first capillary tube through which the first fluid can flow;

(b) a second capillary tube in series with the first capillary tube and through which the second fluid can flow;

(c) a first supply means for supplying die first fluid to flow through bo capillary tubes;

(d) a second supply means for supplying the second fluid to flow through the second capillary tube, comprising a means for introducing a sample material into the first fluid;

(e) means for measuring the pressure drop DP_S across d e first capillary tube and d e pressure drop DP_p across the second capillary tube and generating a signal responsive to each pressure drop;

(f) a signal processor for receiving and processing the ratio of the pressure drop signals of DPp DP_s for use in measuring either said relative viscosity or the intrinsic or inherent viscosity of the sample material independent of the flow rate and temperature fluctuations of the solution and solvent; wherein the improvement comprises

(g) a means for measuring the flowrate of d e flowrate of the first fluid and/or second fluid; and

(h) a flow controller for limiting fluctuations in the flowrate of the first and second fluid.

Preferably, die means for measuring flowrate comprises the means for measuring the pressure DP_S across the first capillary tube and the amplification means. The sample material may comprise, for example, a compound, composition, or solution of a solute. This invention may also be employed wim GPC (Gel Permeation

Chromatography) or other SEC (Size Exclusion Chromatography) analysis means to obtain information on the molecular weight distribution of polymer materials. The invention may also be used as an in-line process monitor or as a stand-alone viscometer. BRIEF DESCRIPTION OF THE DRAWINGS

Fig. 1 is a schematic diagram of one embodiment of the viscometer of die present invention.

Fig. 2 is a schematic of another embodiment of die viscometer when used in connection witii GPC analysis. DETAILED DESCRIPTION OF THE INVENTION

The present invention employs two capillary tubes, preferably arranged in series, as described in U.S. Patents No. 4,578,990 and 4,627,271, incorporated by reference herein. A pressure transducer is connected across each capillary to monitor simultaneously the pressure drop of fluid flowing through each tube. As described in this invention, the pressure drop measured across the first, or reference, capillary tube will correspond to a reference fluid, such as a solvent and will be designated as DP_S. The pressure drop across the second, or sample, capillary tube will correspond to a sample fluid, such as a polymer solution, and will be designated as DP_p. These measured pressure drops, DP_S and DP_p, are men used to determine the inherent or intrinsic viscosity of the sample polymer and to determine the flowrate of the fluids.

While e present invention is described herein with reference particularly to polymer-solvent solutions, it should be understood that the invention may be used with other sample and referencing liquids wherever accurate viscosity measurements are desired. Thus, die viscosity of other sample liquids may be readily calculated from the relative viscosity measurement where the viscosity of me referencing liquid is known. The signals generated by the pressure transducers are fed to an analogue or digital signal processor, alternatively referred to as an amplifier. The signal processor can be an analog divider, but is preferably a high speed computer. In the process and apparatus of this invention, the signal processor processes the pressure drop differentials in two ways: (1) as a ratio DPp/DP_s and (2) as a flowrate based, for example, on DP_S.

The signal processor can perform these functions in real time. Real time signal processing of die simultaneous pressure drops across die analytical and reference capillaries assists in eliminating die effects of flow rate and temperature fluctuations in the capillaries. It also eliminates the need to match die dimensions of d e capillaries. The output of the signal processor is then a real time measure of die natural logarithmic of the relative viscosity, In η_r. The pressure measurement from the reference transducer can also be a real time measure of flowrate. The signal processor can also store the calculated variables for post-processing.

The pressure drop DP signal generated by die pressure transducer is related to the viscosity of fluid flowing through the capillary in accordance widi the following relationship:

DP = KQη where

K is the instrument constant which is proportional to the capillary's length, 1, and diameter, d, as follows: 1/d⁴

Q is die volume flow rate η is the effluent liquid viscosity. After the pressure drop DP signal is received by die signal processor, a net signal (S) can be determined based on the following relationships: (S) = lnDPp -.lnDP_s = ln(DPp/DP_s) = ln(G₁K₁Qη_p/G₂K₂Qη_s)

^■ lnCGiK.ηp/GjKaη,) = ln(G₁K₁/G₂K₂) + ln(η_r η_s) = ln(G₁K₁/G₂K₂) + lnη_r where, as noted earlier h_r is e relative viscosity T|p/η_s and G_j and G2 are the electronic gains in the DP signal, measured across die analytical and reference capillary, respectively. The first term is a flow rate and temperature independent instrument constant of the capillary mismatch that can easily be eliminated by suitable zero offset calibration in the instrument electronics.

As stated above and shown in Figure 1, die solvent and sample capillaries are preferably arranged in a series fashion. It has now been found that controlling the flowrate of d e solvent and sample fluid flowing through the reference and sample capillary tubes, respectively, can very significandy and surprisingly increase the precision of the viscosity measurements obtained. The flowrate control system employs a pressure measurement to determine the flowrate at a given point in combination with a controller which compares the measured flowrate to a predetermined setpoint and then determines the correct adjustment to the flow. This adjustment would be transmitted to die pump and the pump speed would be adjusted accordingly. The flowrate can men be maintained within a predetermined limited range. Since the solvent and sample flow are provided by one pumping means, and the solvent and sample are arranged to flow through their respective capillaries simultaneously, the flowrate of the solvent and sample are essentially equivalent. The flowrate, dierefore, need only be measured at one point. The pressure measurements from the reference pressure transducer, DP_S, is preferably employed for this purpose, since the material flowing through the capillary would not change and die transducer does not experience hydraulic upsets due to sample introduction.

Signals from the pressure transducers are received by the signal processor, as discussed above. A suitable pressure transducer is me Validyne^<!DP15-32-22- 66 differential pressure transducer commercially available from Validyne Corp. (Northridge, CA). The signal processor is preferably micro-processor based, capable of manipulating the pressure measurement to obtain a flowrate. A personal computer suitably modified may be employed as die signal processor, adapted with an A/D board such as Data Translation Corp. Model DT 2800/5716A A/D board. The calculated flowrate is then transmitted to a flow controller. The amplifying means may transmit signals pneumatically, electronically, or digitally, depending on the configuration of the recipient controller. Many flow controllers are known in the art. Any flow controller which serves to control the rate of the solvent and polymer within about five percent of a given setpoint is suitable for use in the invention herein. Various types of controllers can be used. The controller receives die flowrate measurement from the amplifier, compares it with a pre-determined setpoint, and determines the needed adjustment to d e pump speed. The needed adjustment to me flowrate may be determined in various ways. Controllers may use model-predictive control or proportional-plus-integral control schemes, among others, to determine d e necessary adjustment to d e pump speed. Suitable control software is commercially available, for example, Asyst®version 4.1 software commercially available from Kiedily Instruments Corp. A Maxon Corp. MMC^Linear servor controller can be employed to adjust d e pump speed.

Figure 1 illustrates one embodiment of the viscometer of the present invention which can be used in batch, or if a fractionating column is used to supply die sample capillary tube, continuous, sample viscosity determinations. Two capillaries are arranged in series. The first is the reference capillary 10 through which only solvent flows. The second is die sample capillary 12 through which the solute-solvent solution will flow. As one example, this solute may be a polymer. The capillaries are long tubes of small internal diameter formed of glass, metal or any odier suitable material. Connected across each respective capillary are pressure transducers 14 and

16 which monitor the pressure drops for the fluid flowing through the capillaries. Each transducer will generate an electrical signal corresponding to me pressure drop across its capillary. These signals are fed to a signal processor 17.

Solvent is introduced from a reservoir 18 via a pump 20. The pump speed is controlled by a flow controller 19 which adjusts die speed based on input from the process controller 17. The pump can be of various types but is preferably a gear pump. When a relatively pulseless gear pump is used with feedback flow control, a constant, pulseless, flowrate of fluid results.

The polymer sample is injected into die solvent stream from a sample loop 22 via a sample injection valve 24, which may be a two-position 6-port valve. This type of valve is sold by Valco Instruments Inc., under the designation Valco CV6UHPA. In this case, the sample loop is located downstream from the reference capillary 10 but before the sample capillary 12. Solvent is pumped to the reference capillary. A flow resistor 32 provides a solvent flow by-pass around the sample injection valve 24. This ensures continuous flow during sample injection and reduces flow rate upsets caused by die valve switching during sample injection. A concentration detector 34, such as a differential refractometer is placed at the end of die stream. Other types of concentration detectors such as ultraviolet or infrared devices may be used depending upon die particular type of sample used. After it passes die concentration detector 34, the sample stream empties into a waste receptacle 36. Where die concentration of d e sample is known, a concentration detector is not needed in order to calculate either die inherent or intrinsic viscosity of me polymer.

In operation, the viscometer of Figure 1 will generate two separate signal detector traces for recording. The signals from pressure transducers 14 and 16, by employing a signal processor such as an analog divider or high speed computer, will be processed to generate a viscosity (lnη_r) trace. The signal from pressure transducer 14 will also be processed by die signal processor to generate a direct indication of flowrate. This flowrate is fed to die flow controller 19 and is used to determine needed adjustments to die pump speed. The concentration detector 34 will generate a concentration trace. The signals from the pressure transducers can be processed in real-time or stored for further processing at a later time.

Figure 2 illustrates a second embodiment of the viscometer of the present invention where like elements are designated by the same reference numerals. The viscometer here is used as a GPC-viscosity detector, but it should be understood that other separation devices could also be used. A GPC column set 40 is located between die sample loop 22 and injection valve 24 and d e sample capillary 12. It should be noted that the relative positions of the sample (for example, polymer solution) capillary 12 and reference (solvent) capillary 10 have been reversed in this embodiment. A large depository column 42 has been added and die concentration detector 24 is located between the analytical and reference capillary. The purpose of the column 42 is to dilute the polymer-solvent solution with solvent so tiiat die reference capillary 10 essentially has only solvent passing through it during die detection operation.

While the operation of the GPC-viscosity detector will be described with reference to the schematic of Figure 2, it should be understood d at die viscometer of Figure 1 could have been used also, modified only by adding a GPC or other SEC column set between the sample injection loop 22 and die sample capillary 12. Likewise, widi the arrangement of Figure 1, column 42 is not necessary. In one embodiment of die present invention, a multicomponent polymer mixture sample is injected through sample loop 22 and valve 24. The sample is carried by the solvent and is introduced into the GPC column 40 where the polymer molecules are separated according to size. The large molecules are eluted first at die other end as described earlier. The concentration detector 34 detects each polymer component and provides a concentration signal at the elution time corresponding to each successively smaller molecular weight polymer component. The DP viscosity detection traces will also occur at the corresponding retention time since each component will generate a pressure drop signal across the sample capillary 12 as it is eluted from die GPC column set 40.

The DP signals are fed to a signal processor or amplifier, preferably a computer, which will process them as explained above. The DP_S signal is fed to the computer or other signal processor, and resulting flowrate measurements are fed to die flow controller 19. The foregoing also improves the independence of die present invention to temperature fluctuations. As noted earlier, temperature control is important in conventional capillary viscometers since small temperature variations in d e liquids can gready effect the accuracy of the viscosity measurements. By controlling die flowrate in the apparatus under conditions of varying back pressure in die system, the invention herein provides a much more stable heat flux experience for each parcel of fluid as it passes dirough the system, and thus produces more nearly identical temperatures for d e fluids in me sample and reference zones of the instrument. No special temperature controls are necessary with the present invention, although diey may be optionally employed if desired. While several embodiments and applications of die present invention have been shown and described, it would be apparent to tiiose skilled in die art that many more modifications and applications are possible without departing from d e invention, die scope of which is defined by d e claims.

EXAMPLES 1-12 Relative viscosity measurements were taken, using the apparatus shown in

Figure 1, for various Viton^erflouroelastic polymers. For each specific polymer, all measurements were taken from the same jar of emulsion which was taken in one sample from a polymer process. For each polymer, all the measurements were also taken on the same day. The measurements are shown in Tables I-A and I-B. The measurements yielded an average relative standard deviation,

RSD = 0.0061, and a standard deviation, SD = 0.0028. These results indicate an 88% improvement in the standard deviation, and about 63% improvement in the relative standard deviation, over measurements taken without flow control (see Comparative Examples A-M), representing a very significant improvement in the precision of the measurements.

TABLE I-A

PREQSIO N OF RELATIVE V 1SCOSIT i MEASU REMENT 'S

POLYMER 1 2 3 4 5 6

0.528 0.694 0.465 1.765 0.876 0.836

Ηr 0.526 0.693 0.462 1.745 0.862 0.829

Measurements 0.524 0.695 0.471 1.739 0.856 0.834

0.523 0.697 0.472 1.744 0.860 0.835

0.525 0.472 1.744 0.856 0.826

0.527 0.473 1.749 0.867 0.830

0.870

0.866

0.865

Average 0.5255 0.6948 0.4692 1.7477 0.8642 0.8317

SD 0.0019 0.0017 0.0045 0.0091 0.0065 0.0039

RSD 0.0036 0.0024 0.0096 0.0052 0.0075 0.0047

TABLE I-B

PREQSIO N OF RELATIVE VISCOSITY MEASUREMENTS

POLYMER 7 8 9 10 11 12

1.827 0.677 0.942 0.609 0.898 0.677 η_r 1.785 0.677 0.924 0.610 0.894 0.680

Measurements 1.784 0.678 0.922 0.610 0.903 0.680

1.780 0.688 0.923 0.611 0.897 0.678

1.784 0.685 0.918 0.606 0.896 0.688

1.786 0.688 0.927 0.607 0.895

1.782 0.676 0.911

Average 1.7897 0.6813 0.9239 0.6088- 0.8972 0.6806

SD 0.0165 0.0055 0.0095 0.0019 0.0032 0.0043

RSD 0.0092 0.0081 0.0103 0.0031 0.0036 0.0063

COMPARATIVE EXAMPLES 13-25 These examples were run in the same manner as Examples 1-12, however no feedback flow control was used. Each set of measurements of a particular polymer were taken from the same jar of emulsion taken as one sample from a polymer process. Additionally, all d e measurements for each polymer were taken on the same day. The resulting measurements are shown in Tables II-A and II-B.

The measurements yielded an average standard deviation of 0.0224, and an average relative standard deviation of 0.0166.

TABLE II-A PRECISION OF RELATIVE VISCOSITY MEASUREMENTS

POLYMER 13 14 15 16 17 18

1.735 0.531 0.853 0.902 0.785 1.673

1.746 0.479 0.855 0.898 0.841 1.679

Measurements 1.736 0.517 0.846 0.894 0.818 1.684

1.745 0.502 0.856 0.892 0.834 1.689

1.751 0.533 0.853 0.888 0.827 1.693

0.461 0.852 1.694

0.432

Average 1.7426 0.4936 0.8525 0.8948 0.821 1.6853

SD 0.0069 0.0379 0.0035 0.0054 0.0219 0.0083

RSD 0.0039 0.0768 0.0041 0.0060 0.0267 0.0049

TABLE II-B

PRECΪSK 3N OF RELATIVE VISCOSITY MEASUREMENTS

POLYMER 19 20 21 22 23 24 25

0.813 0.820 0.795 0.602 0.583 0.679 0.688

0.800 0.818 0.787 0.601 0.583 0.673 0.680

Measurements 0.799 0.821 0.788 0.614 0.578 0.686 0.676

0.799 0.811 0.790 0.547 0.574 0.680 0.671

0.804

Average 0.8032 0.8175 0.790 0.591 0.5795 0.6795 0.6788

SD 0.0053 0.0045 0.0036 0.0299 0.0044 0.0053 0.0072

RSD 0.0066 0.0055 0.0046 0.0506 0.0076 0.0078 0.0106

Claims

CLAIMS What is claimed is:

1. A method for measuring the relative viscosity of two fluids, a first fluid and a second fluid, which method comprises d e steps of: (a) passing d e first fluid through a first capillary tube;

(c) measuring separately the pressure drop, DP_S, across the first capillary tube and the pressure drop, DP_p, across the second capillary tube when each, respectively, is full of flowing first fluid and second fluid;

(d) generating signals corresponding to die pressure drop across each capillary tube; and

(e) feeding said signals to an analogue or digital signal processor which logarithmically processes the ratio of DPp/DP_s for use in measuring the relative viscosity, wherein die improvement comprises:

(f) determining the flowrate of the flowing first and second fluid by processing a pressure drop signal in the signal processor, and

(g) transmitting a flowrate signal to a controller for limiting fluctuations in the flowrate of the solvent medium and solution.

2. The metiiod of Claim 1 wherein the signal processor processes the ratio of DPp/DP_s in real time.

3. The method of Claim 1 wherein flowrate in step (f) is determined by processing the pressure drop, DP_S, across the first capillary tube.

4. The method of claim 1 wherein the first fluid is a solvent medium and the second fluid is a solution of a solute in ie solvent medium.

5. The metiiod of Claim 4 wherein die solute is a polymer.

6. The mediod of Claim 1 wherein a measure of the inherent or intrinsic viscosity is generated by the signal processor.

7. The method of Claim 1 further comprising the steps of measuring the concentration of d e solute in the solution passing through the second capillary tube and generating a signal corresponding to die solute concentration.

8. An apparatus for measuring the relative viscosity of two fluids, a first fluid and a second fluid, comprising in combination:

(a) a first capillary tube through which the first fluid flows; (b) a second capillary tube arranged in series witii the first capillary tube and through which the second fluid flows;

(c) a first supply means for supplying the first fluid to flow through both capillary tubes; (d) a second supply means for supplying the second fluid to flow through the second capdlary tube, comprising a means for introducing a sample material into the first fluid;

(e) means for measuring die pressure drop DP_S across the first capillary tube and die pressure drop DP_p across the second capillary tube and generating a signal responsive to each pressure drop;

(f) a signal processor for receiving and processing the ratio of the pressure drop signals of DPp/DP_s for use in measuring either said relative viscosity or the intrinsic or inherent viscosity of a solute in d e sample material, wherein the improvement comprises:

(g) a means for measuring the flowrate of the solvent medium and solution, and

(h) a flow controller for limiting fluctuations in die flowrate of the first and second fluid.

9. The apparatus of Claim 8 wherein the means for measuring die flowrate comprises, in combination, the means for measuring the pressure drop DP_S across die first capillary tube and d e signal processor.

10. The apparatus of Claim 9 wherein the means for measuring pressure drop across each capillary tube comprises a pressure transducer.

11. The apparatus of Claim 8 wherein the signal processor is adapted for processing of the ratio of DPp/DP_s in real time.

12. The apparatus of Claim 8, wherein the signal process is adapted for measuring the inherent or intrinsic viscosity of die solute.

13. The apparatus of Claim 8 , further comprising means for measuring the concentration of the solute in die second fluid passing through the second capillary tube and generating a signal corresponding to d e solute concentration.

14. A method of determining the molecular weight distribution of a polymer material as it is eluted from a size exclusion chromatographic column means by a solvent, which method comprises the steps of: (a) passing solvent through d e size exclusion chromatographic column means and through a first and second capillary column arranged in series;

(b) introducing a sample of the polymer material into die solvent before said column means so mat a solution of d e polymer material in solvent flows into the column means; (c) passing die solution through die first capillary tube after it is eluted from said column means;

(d) measuring the pressure drop, DP_S, across the second capillary tube when it is full of flowing solvent, (e) measuring d e pressure drop, DP_p, across the first capillary tube as the polymer material eluted from said column means passes through the first capillary tube;

(f) measuring the concentration of the polymer material as it is eluted from said column means;

(g) generating signals corresponding to the pressure drop and concentration measurements;

(h) feeding said signals to an analogue or digital signal processor which processes the ratio of DPp/DP_s for use in measuring at least the relative viscosity of the solution to the solvent or the intrinsic or inherent viscosity of the polymer material as it is eluted from the column means; and

(i) determining, by means of the signal processor, the molecular weight distribution of die polymer material, wherein the improvement comprises: (j) determining the flowrate of the flowing solvent and solution by processing a pressure drop signal in the signal processor, and

(k) transmitting die flowrate to a controller for limiting fluctuations in d e flowrate of the solvent and solution.

15. The metiiod of Claim 14 wherein the signal processor processes the ratio of DPp/DP_s in real time.

16. The method of Claim 14 wherein flowrate in step (j) s determined by processing the pressure drop, DP_S, across die second capillary tube.

17. An apparatus for determining the molecular weight distribution of a polymer material as it is eluted from a size exclusion chromatographic column means by a solvent, which apparatus comprises in combination:

(a) first and second capillary column arranged in series, wherein solvent flows through the second capillary tube and a solution of a polymer material flows through the first;

(b) solvent supply means for supplying the solvent to flow through both capillary tubes and die column means;

(c) solution supply means for supplying die solution, comprising a means for introducing a sample of a polymer material, the solution supply means located at d e entrance of said column means so that a solution of die polymer material in solvent flows into the column means and die first capillary tube located at die exit of d e column means;

(d) means for measuring the pressure drop, DP_S, across the second capillary mbe when it is full of flowing solvent, (e) means for measuring the pressure drop, DP_p, across the first capillary tube as the polymer material eluted from said column means passes through die first capillary tube;

(f) means for measuring the concentration of the polymer material at it is eluted from said column means;

(g) means for generating signals corresponding to die pressure drop and concentration measurements;

(h) a logarithmic analogue or digital signal processor for processing d e ratio of DPp/DP_s and means for processing d e concentration signal, measuring the relative viscosity of the solution to the solvent or the intrinsic or inherent viscosity of d e polymer material as it is eluted from die column means, preliminary to determining the molecular weight distribution of die polymer material, wherein the improvement comprises:

(i) a means for determining d e flowrate of die flowing solvent and solution by processing a pressure drop signal in die signal processor, and (j) a controller for receiving an input signal based on the flowrate and, based diereon, limiting fluctuations in the flowrate of the solvent and solution.

18. The apparatus of Claim 17 wherein d e means for measuring the flowrate comprises, in combination, the means for measuring the pressure drop

DP_S across the first capillary tube and die signal processor.

19. The apparatus of Claim 17 wherein the means for measuring pressure drop across each capillary tube comprises a pressure transducer.