Application for Energy and Fuels

Application for Energy and Fuels Improved Throughput for Oils Analysis by ICP-OES Using Next Generation Sample Introduction Technology Patrick Simmons, Doug Shrader & Phil Lowenstern Agilent Technologies 1

Overview Application Considerations Applications Biodiesel Petroleum diesel and coolant analysis Naptha Wear metals High Throughput Analysis Conclusions 2

Application Considerations The challenge: High vapour pressure from solvents Plasma instability Extinguished plasma Carbon build up on injector Poor precision and drift Down time - injector requires regular cleaning Nebulizer blockage Poor precision Down time - nebulizer needs cleaning 3

Application Considerations The solution: Carbon build up Use radially viewed vertically oriented plasma Minimizes carbon build-up Exhaust positioned directly above plasma, efficiently extracts carbon by-products Use oxygen addition into auxiliary gas flow to facilitate removal of carbon and reduce background Mandatory with axial ICP Nebulizer blockage Use nebulizer with large ID to reduce blockage from particulates e.g. SeaSpray, Slurry or V-groove nebulizer Use new OneNeb nebulizer Filter samples or allow to settle and sample from the top Conikal nebulizer produces finer aerosol 4

One Piece Torches Radial ICP Torches Designed specially for organic solvents Annealed for greater durability Choice of 0.8 and 1.4mm ID. injector 1.4 mm ID suitable for organic solvents of low to moderate volatility 0.8 mm ID suitable for highly volatile organic solvents 5

Nebulizers Sample particulates may block standard concentric nebulizers Dependent on sample composition e.g. wear metal particulates Poor precision Down time - nebulizer needs cleaning Quartz nebulizer Conikal, Seaspray and Slurry Faster washout V-groove nebulizer Made of PEEK polymer Minimizes particulate blockage Quartz nebulizer V-groove nebulizer 6

For Challenging Applications OneNeb Nebulizer Robust PFA and PEEK construction Inert - resistant to strong acids such as HF Resistant to breakage Molded plastic design provides improved nebulizer to nebulizer reproducibility Constant diameter narrow bore tubing through to nebulizer tip Ideal for high solids/particulates Improved tolerance to high TDS samples Narrow aerosol size distribution provides improved precision Handles a wide flow range from 0.1 to 2 ml/min. No sensitivity loss at low flow rates 7

Principle of Operation Inert OneNeb Nebulizer Tip geometry dimensioned to allow carrier gas to mix with the sample Turbulent mixing of liquid and gas occurs Produces aerosol with a narrow size distribution range Droplets mixed into gas flow No area of low pressure Unique Flow Blurring action increases nebulization efficiency Greater efficiency than conventional concentric nebulizer Improved sensitivity Proportion of Total 14 12 10 8 6 4 2 Inert OneNeb nebulizer Conventional concentric nebulizer 0 0 20 40 60 80 100 Droplet Size ( m) 8

Spray Chambers Double-pass glass cyclonic spray chamber Double pass design Reduces solvent load Increased sensitivity Fast washout Suitable for low vapour organic solvents Cooled spray chamber Externally jacketed or Peltier cooled designs Can cool sample to -10 o C or lower Reduces solvent load Made of quartz Excellent for highly volatile organic solvents e.g. naphtha 9

Compatibility of Pump Tubing Pump Tubing Solvent PVC Viton PVC Solva Kerosene Gasoline Fuel Coolant N/A = unsatisfactory, = satisfactory, N/A = no data available 10

Application Considerations The challenge: Various elements of interest May need to calibrate for 30 to 60 different elements Unknown concentration of analytes in samples Down time to perform multiple dilutions for the same sample Difficult interferences to overcome Poor accuracy 11

Measuring Petroleum Products Successfully The solution: Linear dynamic range Use wavelength flexibility to extend the upper limit Still achieve best detection limits Structured backgrounds Radial ICP-OES Use patented Fast Automated Curvefitting Technique (FACT) Speed of analysis 725 simultaneous CCD system Fastest possible read out speed 12

700 Series Optical Design Echelle Optical Design Full wavelength coverage Maximum flexibility for extended linear working range Elimination of spectral interferences All wavelengths captured in one reading Maximum speed and productivity Fewer optical components Excellent signal-to-noise Lowest detection limits Thermostatted to 35 o C Excellent long term stability Fast start-up, increasing sample throughput 13

World s Most Innovative Detector 725 Patented custom-designed CCD Continuous Wavelength coverage 167 785 nm Simultaneous measurement Very high Quantum Efficiency Excellent resolution Lower detection limits Dual mode anti-blooming capabilities Eliminates blooming in both the X and Y direction Peltier cooled for lowest noise and best detection limits 14

Structured Background Introduction of organic solvents into the ICP can result in spectral interferences from molecular emission by species such as C 2 and CN Na 589nm in engine oil and diluted in white spirit Blank 0 mg/kg Na Standard 10 mg/kg Na 15

Structured Background from Na 589 nm in Oil - Diluted in white spirit Blank 0 mg/kg Na - No Background Correction Blank 0 mg/kg Na Fitted Background Correction 16

Structured Background from Na 589nm in Oil - Diluted in white spirit FACT Corrected Background Sample Spectrum Analyte Blank 0 mg/kg Na Standard 10 mg/kg 17

18 Structured Background from Na 589nm in Oil - Diluted in white spirit FACT Corrected

Sample and Standard Preparation Sample Preparation Up to 1 in 10 dilution on a weight per volume basis Use suitable organic solvent e.g. xylene, kerosene, Shellsol Standard Preparation Multi-element organometallic standard (e.g. Conostan S-21) Select dilution ratio to achieve required concentration Add extra neutral base oil (No. 75) to ensure consistent viscosity 19

Typical Instrument Conditions Plasma power 1.3-1.5kW Plasma gas flow 15 18 L/min Auxiliary gas flow 0.75 2.25 L/min (axial radial) Nebulizer gas flow (axial) Optimize (0.5 0.8 L/min) Nebulizer gas flow (radial) Set bullet to top of torch (0.5 0.8 L/min) AGM-1 setting 2-6 Stabilization delay Optimize via time scan Pump speed 5-10 rpm Uptake Rate < 1 ml/min Fast Pump Yes or No? 20

Important Parameters Parameters: Power 1.3-1.5 kw Nebulizer flow: optimize using bullet of the plasma Viewing height: optimized for SBR Each can be automatically optimized using AutoMax 21

Why Optimize? Power = 1.2 kw Nebulizer flow = 1.00L/min Viewing height = 10mm Power = 1.5 kw Nebulizer flow = 1.70L/min Viewing height = 5mm 22

23 Performance Test Linear Range Radial, S21 Conostan Oil Standards in Kerosene

ICP-OES Accessories Overview SPS-3 = Autosampler for automated sample presentation AGM-1 = Addition of oxygen to plasma ASA = Increases the water vapor in the nebulizer gas SVS-2 = Increased productivity and more efficient washout VGA-77P = for hydride determinations of As, Se & Hg (ppb levels) 24

25 Direct Biodiesel Fuel Analysis

Why Use Biodiesel? Reduced dependence on imported oil products Direct replacement for conventional petroleum based diesel fuel Important alternative fuel source Biodiesel is a green fuel Produced from domestic, renewable resources Biodegradable and non-toxic Reduction in pollutants such as sulfur and aromatics 78 % reduction in CO 2 emissions compared with petroleum 26

Analytical Results ASTM D6751 Specified Elements in Biodiesel Element P S Na K Ca Mg Wavelength (nm) 213.618 181.972 589.592 766.491 393.366 279.553 B 100a ppm < IDL 1.695 0.162 0.507 0.013 0.011 B 100b ppm < IDL 1.710 0.130 0.469 0.012 0.012 B 100+1.0 ppm (%R) 105 103 104 107 105 104 B 100+2.5 ppm (%R) 102 106 102 104 103 101 IDL (ppb) 19.9 76.5 4.1 21.8 0.1 0.1 Bkgd Correction Fitted Fitted FACT FACT Fitted Fitted 27

Direct Measurement of Petroleum Diesel and Coolant Samples 28

29 Detection Limits Petroleum Diesel

Stability Petroleum Diesel 1.2 1 0.8 0.6 0.4 0.2 18:56:39 19:13:42 19:30:44 19:47:47 20:04:52 20:21:56 20:39:00 20:56:05 21:13:10 21:30:14 21:47:20 22:04:25 22:21:32 22:38:38 22:55:45 30 0 Ag 328.068 Al 396.152 B 249.772 Ba 455.403 Ca 396.847 Cd 214.439 Cr 267.716 Cu 327.395 Fe 238.204 Mg 279.553 Mn 257.610 Mo 202.032 Ni 231.604 P 213.618 Pb 220.353 S 180.669 S 181.972 S 182.562

Plasma Stability Concentration (ppm) 10 9 8 7 6 5 4 3 2 1 8 Hour Stability Run - NO Int. Std. Worksheet S21Stab7.vws 0 0:00 1:12 2:24 3:36 4:48 6:00 7:12 Time (Hrs:Min) Ag 328.068 Al 396.152 Ba 455.403 Ca 396.847 Cd 214.439 Cr 267.716 Cu 327.395 Fe 238.204 Mg 279.553 Mn 257.610 Mo 202.032 Na 588.995 Na 589.592 Ni 231.604 P 213.618 Pb 220.353 Si 251.611 Sn 189.927 Ti 336.122 V 292.401 Zn 213.857 90 % Limit 110 % Limit Eight hour stability run for elements in oil/kerosene - Conostan S21 31

Determination of Trace Elements in Naphtha Michelle Morin, Natural Resources Canada Wayne Blonski, Agilent Technologies, Mississauga, Ontario, Canada 32

Why Determine Trace Elements in Naphtha? Naphtha is a liquid intermediate between the light gases in the crude oil and the heavier liquid kerosene and are: Volatile Flammable, and Have a low specific gravity of about 0.7 The presence of trace metals in hydrocarbon samples can: Severely hamper the catalytic reforming process Poison the catalysts used e.g. S and N compounds can deactivate catalysts Catalysts used in catalytic cracking or reforming Photo courtesy Phillips Petroleum Company 33

Performance Test - Plasma Stability 7 Concentration 6.5 6 5.5 5 4.5 4 3.5 Ni 231.604 Ni 216.555 Mg 279.553 V 309.310 V 292.401 Mg 280.270 Al 396.152 Ca 396.847 Ca 422.673 Cd 214.439 Cd 226.502 Cd 228.802 Cr 267.716 Cu 324.754 Cu 327.395 Fe 238.204 Fe 259.940 Mn 257.610 Na 588.995 Na 589.592 3 1 : 1 1 : 2 1 : 3 1 : 4 1 : 5 1 : 6 1 : 7 1 : 8 1 : 9 1 : 10 1 : 11 1 : 12 1 : 13 1 : 14 1 : 15 1 : 16 1 : 17 1 : 18 1 : 19 1 : 20 No. Samples Measured Short term stability over 30 mins. for Naphtha 34

35 Detection Limit Performance (mg/l)

Wear Metals Analysis A High Throughput Application 36

Accuracy CRM Analysis 1084a Wear metals in lubricating oil Elements & Wavelength (nm) Results (ppm) CRM 1084a (ppm) % Recovery Ag 328.068 96.3 101.4 95 Al 167.019 105.6 104 102* Cr 267.716 96.9 98.3 99 Cu 327.395 99.1 100 99 Fe 238.204 100.6 98.9 102 Mg 279.553 100.6 99.5 101 Mo 202.032 96.6 100.3 96 Ni 231.604 99.5 99.7 100 Pb 220.353 107.7 101.1 107 S 181.972 2022 1700 119** Si 251.611 100.7 103 98* Sn 189.927 91.5 97.2 94 Ti 336.122 101.3 100.4 101 V 292.401 101.3 100.4 101 * Uncertified results ** Uncertified results, result was high because of kerosene contamination 37

Introduction In order to predict when equipment maintenance may be required or to prevent having to perform maintenance, lubricating oils in equipment are regularly analyzed to monitor changes in levels of wear-metals, and additive and contaminant elements. The analyst is mainly interested in trending changes over time, not exact values. So high sample throughput could be considered more important than accuracy, precision, long-term stability and repeatability/reproducibility. In this work, an Agilent 725 Series ICP-OES inductively coupled plasma optical emission spectrometer with Agilent SPS 3 Sample Preparation System and Agilent SVS 2 Switching Valve System was used. The SVS 2 improves efficiency by greatly reducing sample uptake and washout times. The typical ICP-OES sample analysis cycle time was halved (to 33 seconds per sample), significantly reducing operating costs, without compromising accuracy, precision, long-term stability and repeatability/reproducibility. 39

Measurement challenge Challenge While long-term stability and repeatability/reproducibility are important in wear-metal analysis, since analytical results are used only for trend analysis, accuracy becomes a less important factor and sample throughput is often the most critical consideration. Solution Using the Agilent SVS 2 Switching Valve System with an Agilent 725 Series radiallyviewed ICP-OES and Agilent SPS 3 Sample Preparation System more than halves the sample analysis cycle time of about 90 seconds without the SVS 2, to about 33 seconds per sample using the SVS 2, without compromising accuracy, precision or stability. 40

Experimental Instrumentation Instrument: Agilent 725 Series simultaneous ICP-OES with radially-viewed plasma. Accessories: Agilent SPS 3 Sample Preparation System; Agilent SVS 2 Switching Valve System. 725 ICP-OES Switching Valve System 2 Increase Productivity Higher sample throughput Decreased wear and tear on nebulizer/spraychamber/torch Lower operating cost through reduced argon use 41

SVS Productivity Packages How does it work? Comparison of the sample uptake, measurement and rinse profiles of 100 mg/l manganese without a SVS and the SVS 2. The data shows a dramatic increase in productivity while maintaining consistent data quality. 42

SVS 2 Switching Valve System Second Generation Switching Valve SVS 2 in Sample Load Position SVS 2 in Sample Inject Position Load Analyze 43

Experimental Instrumentation Agilent 725 Series ICP-OES instrument operating parameters Condition Setting Power 1.35 kw Plasma gas flow rate 15 L/min Auxiliary gas flow rate 2.25 L/min Spray chamber Glass cyclonic double-pass (Twister) Torch One-piece quartz radial (1.5 mm id injector) Transfer tube Glass Nebulizer Glass concentric (SeaSpray) Nebulizer flow rate 0.55 L/min Viewing height 9 mm Pump tubing Rinse/instrument: Gray/gray SolventFlex (1.30 mm id) Waste: Purple/black SolventFlex (2.29 mm id) Pump speed 12 rpm Total sample usage 2 ml Replicate read time 2 s Number of replicates 3 Sample uptake delay 0 s Stabilization time 12 s Rinse time 0 s Fast pump Off Background correction Fitted Note: An all-glass sample introduction system (part number 9910117900) was used. 44

Experimental Instrumentation Agilent SVS 2 Switching Valve System operating parameters Condition Loop uptake delay Uptake pump speed refill Uptake pump speed inject Sample loop size Time in sample Bubble inject time Setting 7 s 500 rpm 150 rpm 0.5 ml 6 s 6.9 s Note: The internal standard/diluent channel was not used. 45

Experimental Standard and sample preparation Calibration solutions of 0, 5, 10, 25 and 50 mg/l were prepared from Conostan S-21 + K certified standard, which contains 22 elements (Ag, Al, B, Ba, Ca, Cd, Cr, Cu, Fe, K, Mg, Mn, Mo, Na, Ni, P, Pb, Si, Sn, Ti, V and Zn) at 500 mg/kg in oil. These calibration solutions were viscosity matched using Conostan Element Blank Oil (75 cst) and diluted with kerosene to give a total oil concentration of 10 % (w/v) in each solution. Duplicate 0.5 g portions of NIST SRM 1084a (Wear-Metals in Lubricating Oil) and 2 g of Element Base Oil were accurately weighed into 25 ml volumetric flasks and made up to volume with kerosene. A third 0.5 g portion was similarly prepared, spiked with S-21 + K standard. Duplicate 0.5 g portions of NIST SRM 1085b (Wear-Metals in Lubricating Oil) and 4.5 g of Element Base Oil were accurately weighed into 50 ml volumetric flasks and made up to volume with kerosene. 46

Experimental 3σ method detection limits (MDLs) & linearity correlation coefficients Element &wavelength MDL (mg/l) Ag 328.068 0.003 Al 308.215 0.018 B 249.678 0.011 Ba 493.408 0.002 Ca 422.673 0.008 Cd 228.802 0.005 Cr 205.560 0.010 Cu 327.395 0.007 Fe 259.940 0.004 K 766.491 0.081 Mg 285.213 0.006 Mn 260.568 0.004 Mo 204.598 0.015 Na 588.995 0.062 Ni 230.299 0.018 P 177.434 0.069 Pb 283.305 0.031 Si 251.611 0.101 Sn 283.998 0.128 Ti 334.941 0.002 V 292.401 0.003 Zn 213.857 0.015 Element &wavelength r 2 Ag 328.068 0.9998 Al 308.215 0.9998 B 249.678 0.9998 Ba 493.408 0.9996 Ca 422.673 0.9996 Cd 228.802 0.9999 Cr 205.560 0.9999 Cu 327.395 0.9997 Fe 259.940 0.9999 K 766.491 0.9996 Mg 285.213 0.9998 Mn 260.568 0.9998 Mo 204.598 0.9999 Na 588.995 0.9999 Ni 230.299 0.9999 P 177.434 0.9999 Pb 283.305 0.9997 Si 251.611 0.9999 Sn 283.998 0.9999 Ti 334.941 0.9997 V 292.401 0.9999 Zn 213.857 0.9999 47

Results NIST SRM 1084a Wear-metals in lubricating oil sample source NIST US Department of Commerce. Note: Values in parentheses () not certified (information only). Element & wavelength Certified (mg/kg) Found (mg/kg) Duplicate (mg/kg) Recovery (%) RPD dup. (%) Spike level (mg/l) Recovered (mg/l) Recovery (%) Ag 328.068 101.4 100.3 101.1 98.9 0.9 2.17 2.27 104.6 Al 308.215 (104) 99.7 100.4 95.9 0.7 2.17 2.27 104.3 Cr 205.560 98.3 104.3 105.5 106.1 1.2 2.17 2.27 104.3 Cu 327.395 100.0 102.7 103.6 102.7 0.8 2.17 2.27 104.5 Fe 259.940 98.9 105.2 105.4 106.4 0.2 2.17 2.25 103.5 Mg 285.213 99.5 102.5 102.9 103.0 0.4 2.17 2.28 104.9 Mo 204.598 100.3 106.2 106.3 105.9 0.1 2.17 2.28 104.9 Ni 230.299 99.7 106.2 107.0 106.5 0.7 2.17 2.28 104.9 Pb 283.305 101.1 103.1 105.7 102.0 2.5 2.17 2.30 105.9 Si 251.611 (103) 100.2 100.4 97.3 0.2 2.17 2.31 106.1 Sn 283.998 97.2 105.8 105.6 108.8 0.2 2.17 2.31 106.3 Ti 334.941 100.4 105.2 105.2 104.8 0.0 2.17 2.27 104.5 V 292.401 95.9 105.4 105.9 109.9 0.4 2.17 2.30 106.0 48

Results NIST SRM 1085b Wear-metals in lubricating oil sample source NIST US Department of Commerce. Note: Values in parentheses () not certified (information only), values in brackets {} not certified (reference only). Element & wavelength Certified (mg/kg) Found (mg/kg) Duplicate (mg/kg) Recovery (%) RPD dup. (%) Ag 328.068 304.6 307.0 315.9 100.8 2.9 Al 308.215 {300.4} 300.8 306.6 100.1 1.9 B 249.678 (300) 312.4 327.8 104.1 4.9 Ba 493.408 (314) 331.3 339.1 105.5 2.4 Ca 422.673 (298) 292.7 300.6 98.2 2.7 Cd 228.802 302.9 305.1 310.8 100.7 1.8 Cr 205.560 302.9 324.1 329.7 107.0 1.7 Cu 327.395 295.6 303.0 311.1 102.5 2.7 Fe 259.940 {301.2} 310.7 316.9 103.1 2.0 Mg 285.213 297.3 303.6 309.8 102.1 2.1 Mn 260.568 (289) 291.3 296.8 100.8 1.9 Mo 204.598 (296) 312.7 317.8 105.6 1.6 Na 588.995 305.2 299.4 309.7 98.1 3.5 Ni 230.299 295.9 315.5 319.7 106.6 1.3 P 177.434 {299.9} 317.0 317.4 105.7 0.1 Pb 283.305 297.7 308.7 313.2 103.7 1.5 Si 251.611 {300.2} 315.1 314.8 105.0 0.1 Sn 283.998 (294) 317.2 322.7 107.9 1.7 Ti 334.941 {301.1} 311.5 317.2 103.5 1.8 V 292.401 297.8 309.1 314.5 103.8 1.8 Zn 213.857 296.8 308.8 314.9 104.1 2.0 49

Results System stability 50

Experimental Washout / carryover & speed of analysis Washout/carryover Blank measurement Carryover (% of standard concentration) Ba 493.408 Ca 422.673 Fe 259.940 Cu 327.395 Mg 285.213 Zn 213.857 1 0.039 0.047 0.035 0.043 0.045 0.027 2 0.039 0.032 0.032 0.046 0.057 0.026 3 0.052 0.034 0.035 0.039 0.058 0.037 4 0.050 0.034 0.038 0.040 0.047 0.037 Speed of analysis Tube-to-tube analysis time averaged 33 seconds, equating to > 100 samples/hour. 51

SVS 2 Results Summary Were able cut the sample analysis cycle time from about 90 seconds to about 33 seconds per sample using the SVS 2, without compromising accuracy, precision or stability. Dead time is eliminated (sample uptake, stabilization and washout times) SVS 2 utilizes stacked switching valves, sample loop and high speed, positive displacement motor Constant solution flow improves plasma stability Sample never contacts pump tubing, inert sample path reduces sample carryover 52

Application Papers Available on the Agilent Technologies Web site (http://www.agilent.com) SI-A-1413 Determination of metals in oils by ICP-OES SI-A-1417 Determination of V, Ni and Fe in crude oils and bitumen with Sc as an internal standard SI-A-1420 Determination of wear metals in lubricating oil with Axial ICP SI-A-1422 Determination of Pb in Unleaded Gasoline with Axial ICP SI-A-1423 Determination of trace elements in a xylene solution of oil by ICP-AES with ultrasonic nebulization and membrane desolvation SI-A-1427 Multi-element analysis of fuel and lubricating oils by simultaneous ICP-OES SI-A-1431 Improving Throughput for Oils Analysis by ICP-OES PLUS: SI-A-1202, SI-A-1415 and SI-A-1418 53

Questions? THANK YOU 54 Next