Droplet Size Measurement Using Laser Reflection Applications to the Oil and Gas Industry

Droplet Size Measurement Using Laser Reflection Applications to the Oil and Gas Industry PWE Club Meeting Aberdeen December 12 th 2013 Ian Haley, Mettler-Toledo Ian.haley@mt.com +44 (0)7973 859 625

Agenda Overview of laser reflection instrumentation - ParticleTrack FBRM and PVM FBRM evaluation at TUVNEL November 2011 Optimizing i i Oil-Water Separation Using Inline ElectroCoalescer l - Summary of FMC paper at IPTC

Particle System Characterization METTLER TOLEDO is the world leader for monitoring and measurement of particles and droplets as they naturally exist in process METTLER TOLEDO has over 25 years of innovation in methods for Particle System Characterization. Our global team of application consultants has implemented over 2700 probe based installations ti supporting crystallization, emulsions, suspensions and other particle and multi-phase applications Over 5000 scientists and engineers have used our technology around the world Redmond, Washington, USA

In-Situ Particle Characterization Tools ParticleTrack FBRM Technology FBRM Technology Focused Beam Reflectance Measurement Focused Beam Reflectance Measurement PVM Technology Particle Vision and Measurement Temperature G400 #/sec 0 20µm Time Chord Length (µm) 10 µm droplets

What is PVM? Particle Vision and Measurement Inline imaging of Particles and Droplets: - Real time visualization at full process concentration without sampling or dilution Probe-based microscopy - Characterization and understanding of how particles and droplets directly affect product quality and processing efficiency Easy installation and dependable results - PVM probes are designed for process environments PVM allows us to see and understand particles and particle structures as they naturally exist in process ISO 9001 CE

PVM provides immediate process understanding Particle Vision and Measurement (PVM ) View particles as they exist in process and capture valuable information that was not previously available Benefit from the unique perspective of PVM for a better understanding of your particle or droplet system Understanding that goes beyond measurement See and understand changes in particle and droplet systems in real time. e.g.: oiling out of an API before crystallization ti PVM allows you to see particles and droplets as they naturally exist at full process concentration

Seeing is believing High resolution imaging under extreme conditions Even in opaque and highly viscous fluids Shown: Water droplets in crude oil Expect the unexpected Challenges your assumptions about the process Shown: Polymorph detection Measure even when offline sampling is impossible ibl Sampling can change the particle system and can miss fast process dynamics Shown: Dendritic crystals View particles as they exist in process to capture valuable information that was not previously available

Introduction to FBRM Focused Beam Reflectance Measurement (FBRM ) FBRM is a probe based measurement that tracks the rate and degree of change to particles and particle structures as the particles naturally exist in process Measure particles in process without the need for sampling and off-line analysis Real-time measurement of particle count, dimension and shape from submicron to 3mm FBRM is the standard method for tracking changes in particles and droplets in process and in real time

FBRM measures particles without sampling Focused Beam Reflectance Measurement Precise and sensitive measurements at full process concentrations, in opaque or translucent slurries, and in lab or plant environments 175um 300um Track real-time changes with precision and sensitivity The FBRM Chord Length Distribution provides a fingerprint of the particles and droplets as they naturally exist in process Chord Length (µm) Increase in Fines Increase in Filtration Rate Relate directly to product quality and process efficiency The sensitive FBRM measurement can often be directly correlated to the product quality or process efficiency parameters of interest. t FBRM provides a precise and sensitive measurement of changes in particle count, dimension and shape

The FBRM Method of Measurement Cutaway view of FBRM In-process Probe PVM image illustrating the view from the FBRM Probe Window Laser source fiber Detection fiber Beam splitter Rotating optics FBRM Probe Tube Sapphire Window Probe installed in Probe installed in process stream Focused beam

The FBRM Method of Measurement Enlarged view PVM image illustrating the view from the FBRM Probe Window Path of Focused Beam Probe detects pulses of Backscattered light And records measured Chord Lengths

The FBRM Method of Measurement Enlarged view Path Path of of Focused Beam Beam Thousands of Chord Lengths are measured each second to produce the FBRM Chord Length Distribution :

Weighted vs. Unweighted Distributions An explanation No Weighted Distribution Time=0 min Time=90 min Time=180 min Square weighted Distribution Time=0 min Time=90 min Time=180 min On Left: No Weighted FBRM distributions at 3 time points show a decrease in count and an increase in dimension Unweighted distribution is sensitive to fine particles population On Right: Square Weighted FBRM distribution for same 3 time points show enhanced resolution to growth in coarse particle dimension Square weighted distributions is sensitive to coarse particle dimension Example data shown 12 Mettler Toledo Confidential

FBRM Distributions and Trended Statistics An explanation Unweighted Distribution #/s <50 µm Time=0 min Time=90 min Time=180 min #/s 50-1000 µm Square Weighted Distribution Time=0 min Time=90 min Time=180 min Mean 180 = 141µm Mean 90 = 82µm Mean 0 = 75µm Example data shown 13 Mettler Toledo Confidential

Equipment Used Experiments were carried out using an FBRM G400 14mm probe. - Installed in 1 sampling loop - Mounted at 45 into flow direction. Fluid velocity in sampling loop can be varied. - Most measurements isokinetic - Some sub-kinetic and super-kinetic measurements made Position of sampling loop take-off in main flow line can be varied - Three positions used: top, middle and bottom. Main flow line FBRM G400 Loop take-off Sampling loop 15

Experiments Performed 1 Water cut (%) Main loop flow (ms -1 ) Take-off position Sampling loop flow 1 0.5 Top Isokinetic 1 0.5 Bottom Isokinetic 1 1.0 Top Isokinetic 1 1.0 Bottom Isokinetic 1 20 2.0 Top Isokinetic 1 2.0 Bottom Isokinetic 5 0.5 Top Isokinetic 5 05 0.5 Bottom Isokinetic 5 1.0 Top Isokinetic 5 1.0 Bottom Isokinetic 30 02 0.2 Top Isokinetic 30 0.2 Bottom Isokinetic 30 0.5 Top Isokinetic 30 05 0.5 Bottom Isokinetic 30 1.0 Top Isokinetic 30 1.0 Bottom Isokinetic 30 20 2.0 Top Isokinetic 30 2.0 Bottom Isokinetic 16

Summary of Results 1 Flow velocity 0.2ms -1 0.5ms -1 1.0ms -1 2.0ms -1 Water cut 0% 30% 1% 5% 30% 1% 5% 30% 1% 30% C10 (sq wt) Top 11.5 6.0 3.9 3.4 5.3 3.2 3.6 5.3 4.1 9.0 Bottom -- 89.5 4.0 34.9 39.4 3.8 4.2 29.6 4.2 12.6 Mean (sq wt) Top 18.4 20.3 12.1 11.6 27.2 11.7 23.6 44.3 17.5 35.4 Bottom -- 361.4 35.6 150.8 123.0 23.7 29.9 108.3 17.9 50.1 C90 (sq wt) Top 47.1 38.2 22.5 22.0 61.0 22.8 58.1 88.2 36.6 69.0 Bottom -- 761.9 109.8 293.6 233.2 58.2 66.2 210.2 37.3 98.5 All values in microns All sampling isokinetic Mettler Toledo Confidential

40 4.0 5% Water Cut Data FBRM Normalized Distributions Weighted Data 3.0 2.0 1.0 0.0 1 10 100 1000 Chord Length (microns) METTLER TOLEDO METTLER TOLEDO This slide shows FBRM chord length distributions taken during the test performed with 5% water cut. Data for the different flow rates and top and bottom take-off points is again shown. 350 300 250 200 Unweighted Data Again, at the low flow velocity, there is a large difference between the top and bottom data. This difference narrows as the flow velocity increases. This difference is much more pronounced at 5% water cut than at 1% water cut. 150 100 50 0 1 10 100 1000 Chord Length (microns) METTLER TOLEDO METTLER TOLEDO Mettler Toledo Confidential

30% Water Cut Data FBRM Normalized Distributions Weighted Data 2.0 1.5 1.0 0.5 0.0 1 10 100 1000 Chord Length (microns) METTLER TOLEDO METTLER TOLEDO This slide shows FBRM chord length distributions taken during the test performed with 30% water cut. Data for the different flow rates and top and bottom take-off points is again shown. Unweighted Data Again, at the low flow velocity, there is a large difference between the top and bottom data. This difference narrows as the flow velocity increases. This difference is even more pronounced at 30% water cut than at 1% or 5% water cut. counts (No Weight) Unlike the 1% and 5% data, there is still a noticeable difference at 2ms -1 flow velocity between top and bottom using 30% water cut. Mettler Toledo Confidential

300 250 1% and 5% water cut tests FBRM trends 300 250 7000 6000 200 5000 microns 150 100 counts 200 150 100 co ounts (fines) 4000 3000 2000 50 50 1000 0 0 00:15:00 00:30:00 00:45:00 01:00:00 01:15:00 01:30:00 01:45:00 Relative Time METTLER TOLEDO METTLER TOLEDO This graph shows FBRM time-based trends for the 1% and 5% water cut tests. The FBRM instrument provides a very rapid real-time and continuous output showing how droplet size and droplet population is changing as a function of process variables. Each small orange triangle on the y-axis indicates a manual annotation, showing when changes were made to the flow rate, water cut and sampling position. FBRM is thus an ideal tool for following a dynamic process in real time and quantifying rapidly and precisely, the rate of change and degree of change to droplet population and droplet size. As such, deviations form normal process conditions can be identified and quantified easily and quickly. Mettler Toledo Confidential

Conclusions FBRM G400 has successfully tracked changes to droplet size and droplet number of crude oil-water mixtures. Sensitive and precise data was obtained in real-time, with measurements being made continually for many hours each day. FBRM was able to quantify the degree of inhomogeneity in the main flow line as a function of water cut and flow rate: - Increasing water cut increases the degree of inhomogeneity; - Decreasing flow velocity increases the degree of inhomogeneity; - A homogeneous system was only achievable at water cuts of 5% or lower; - Increasing water cut and/or decreasing flow rate results in larger droplets being present. FBRM was also able to measure and quantify long-term changes to the base oil and was able to track separator efficiency as a function of flow rate. FBRM is thus an excellent tool for understanding, optimising, quantifying and controlling the size and number of water droplets in flowing crude oil streams. Mettler Toledo Confidential

Optimizing Oil-Water Separation FBRM is applied to measure droplet distributions downstream of a compact 4 Inline ElectroCoalescer (FMC Technologies). No sampling or dilution is required for droplet measurements. The InLine ElectroCoalescer (IEC) enhances the separation of the multiphase flow, by increasing the water droplet size. IEC causes frequent collisions between water droplets and an electrical field, which destabilizes the film between droplets which have collided, thereby increasing the probability for coalescence. (offshore installation North Sea, MI SWACO) Optimizing Oil-Water Separation Using the Inline ElectroCoalescer Remko Westra, FMC Technologies, Separation Systems, Delta 101 6825 MN Arnhem, The Netherlands separation@fmcti.com Westra, R.W. et al., International Petroleum Technology Conference 14917.

Optimizing Oil-Water Separation Optimizing Oil-Water Separation Using the Inline ElectroCoalescer 4 Remko Westra, FMC Technologies, Separation Systems, Delta 101 6825 MN Arnhem, The Netherlands separation@fmcti.com FBRM is inserted into the pipe spool just downstream of the IEC, and the real time droplet distribution measurements are used to optimize the performance of the oil-water separation. 24 (offshore installation North Sea, MI SWACO)

Optimizing Oil-Water Separation Optimizing Oil-Water Separation Using the Inline ElectroCoalescer 4 Remko Westra, FMC Technologies, Separation Systems, Delta 101 6825 MN Arnhem, The Netherlands separation@fmcti.com As the emulsion droplet size distribution increases, downstream separators perform better. Real-time droplet count and dimension measurements provide quantitative feedback tracking the effect of changes in the incoming i Bimodal water cut, flow rate, and pressure drop and help identify the optimal electrical field strength to improve downstream water-oil separations 25

Water Cut Dependence Optimizing Oil-Water Separation Using the Inline ElectroCoalescer 4 Remko Westra, FMC Technologies, Separation Systems, Delta 101 6825 MN Arnhem, The Netherlands separation@fmcti.com FBRM data indicates that the droplet growth factor of the IEC increases with increasing water cut due to the increased coalescence probability at higher water cuts. 26

Flow Rate Dependence Optimizing Oil-Water Separation Using the Inline ElectroCoalescer 4 Remko Westra, FMC Technologies, Separation Systems, Delta 101 6825 MN Arnhem, The Netherlands separation@fmcti.com FBRM data indicates that the droplet growth factor of the IEC decreases with increased flow rate (which corresponds to a reduced IEC residence time). 27

Electric Field Strength Dependence Optimizing Oil-Water Separation Using the Inline ElectroCoalescer 4 Remko Westra, FMC Technologies, Separation Systems, Delta 101 6825 MN Arnhem, The Netherlands separation@fmcti.com FBRM data indicates that the droplet growth factor of the IEC increases with increasing electric field strength. This is caused by the increased force acting on the water droplets. 28

Thank you for your attention Do you have any Questions? For further information on applications, products and technology please visit: www.mt.com/fbrm www.mt.com/pvm Ian Haley Senior TAC, Mettler-Toledo AutoChem Tel: +44 (0)7973 859 625 Email: ian.haley@mt.com 29