December, 2006 Revised, December, 2007

Quasi-Elastic Light Scattering

Instrument (Brookhaven)

Standard Operating Procedures Section 1: Laser Safety Section 2: Instrument Operation Section 3: Data Handling

TABLE OF CONTENTS

Section 1: Laser Safety Part 1: Eye Protection Part 2: Laser Hazard Classification Part 3: Laser Safety and the Eye Part 4: Specification Sheet for Melles Griot 25 LHP 928 Laser Part 5: Registration Form for Laser Part 6: Laser Eye Protection Selection Table (from American National

Standard Z136.1) Part 7: Spectral Characteristics of Laser Eyewear Part 8: Specification Sheet for Laser Eyewear D14 Part 9: Sample Certificate for Laser Eyewear D14 Part 10: Specification Sheet for Laser Eyewear KOS-6102

Section 2: Instrument Operation Part 11: Detector System Operation Part 12: Laser Operating Instructions Part 13: Laser Operation and Information Manual Part 14: Packing List Part 15: Dynamic Light Scattering (i.e. QELS) Operating Instructions Section 3: Data Handling Part 16: Simple Introduction to DLS Data Analysis Part 17: Molecular Weight Determination Using Zimm Plot Part 18: QELS Operational Considerations

Section 1

Laser Safety

EYE PROTECTION Do NOT look directly into the laser beam. Wear protective eyewear when working near the laser beam. The protective eyewear must be appropriate for the wavelength and power of the laser beam. The eyewear provided in the lab (B122) is appropriate for this laser- it provides an optical density of at least +5 at 632.8nm. The eyewear specifications are described in the next sections.

Laser Hazard Classification

Quicklinks

● FAQ

● Radioactive Waste Management

● Spill Procedures

● UIUC Radiation Safety Manual

● Survey Procedures

● Open Source Radioactive Material Safety Tutorial

● Radioactive Decay Calculator

● Radioactive Unit Converter

● Sanitary Sewer Discharge Limits Calculator

● Analytical X-ray Safety Tutorial

● Laser Safety Tutorial

Links of Interest

● Office of the Vice Chancellor for Research

● UIUC Campus Administrative Manual

● Claims Management Office

● Division of Public Safety

Search instructions

Laser Hazard ClassificationClasses of Lasers (adopted from ANSI Z-136.1-1993*)

● Class 1 1. Not capable of emitting in excess of the Class 1 AEL 2. Most lasers in this class are lasers which are in an enclosure which prohibits

or limits access to the laser radiation ● Class 2a

1. Lasers in the visible region of the spectrum which do not exceed the Class 1 AEL for exposure less than or equal to 10E3 s

2. The output of the laser is not intended to be viewed 3. An example of a Class 2a laser is a supermarket point-of-sale scanner

● Class 2 1. All Class 2 lasers are in the visible region of the spectrum 2. Continuous wave lasers which can emit accessible radiant power which

exceeds the Class 1 AEL for the maximum duration inherent in the laser, but do not exceed 1 mW

3. Pulsed lasers which can emit accessible radiant power which exceeds the Class 1 AEL for the maximum duration inherent in the laser, but not do not exceed the Class 1 AEL for an exposure of 0.25 s

● Class 3a 1. Have output that is greater than or equal to 5 times Class 2 AELs

● Class 3b 1. Continuous wave - between the Class 3a limits and 500 mW 2. Repetitively pulsed - radiant energy between 30-150 mJ per pulse for visible

and infrared, otherwise greater than 125 mJ per pulse; average power less than 500 mW

● Class 4 ❍ Limits exceed Class 3b limits

Biological Safety | Chemical Safety | Radiation Safety | Home | UIUC | About RSS | Contact Information | Training | Fact Sheets | Rad Safety Manual | Rad Materials | X-Ray | Laser

Information concerning Division of Research Safety programs of the University of Illinois at Urbana-Champaign is intended as guidance for University of Illinois at Urbana-Champaign students, staff, and faculty engaged in activities related to their education, research, and/or employment. The information is subject to change and updating at any time.

Copyright © 2003 University of Illinois at Urbana-Champaign Division of Research Safety.

Contact webmaster.

http://www.ehs.uiuc.edu/rss/laser/laserhazard.htm8/9/2004 10:35:59 AM

Laser Safety and the Eye

Laser Safety and the Eye:Hidden Hazards and Practical Pearls

Osama Bader, MD, and Harvey Lui, MD, FRCPC

From the Lions Laser Skin Centre, Division of Dermatology,Vancouver Hospital & Health Sciences Centre,and University of British Columbia, Vancouver, B.C.

Presented at the American Academy of DermatologyAnnual Meeting Poster Session,Washington, D.C.February 10-15, 1996

Contents

● Abstract● What are the effects of laser energy on the eye?● Are there any specific symptoms of laser eye injuries?● What types of laser safety eyewear are available?● What are the technical considerations for eye safety?● What factors should be considered when selecting specific eyewear?● Practical Pearls in Laser Eye Safety● References

http://www.dermatology.org/laser/eyesafety.html (1 of 7)8/9/2004 10:33:58 AM

Laser Safety and the Eye

● Acknowledgements

(Click the images below for a magnified view.)

Abstract

THE UNPROTECTED HUMAN EYE is extremely sensitive to laser radiation and can be permanently damaged from direct or reflected beams. The site of ocular damage for any given laser depends upon its output wavelength. Laser light in the visible and near infrared spectrum 400 - 1400 nm (the majority of lasers used in dermatology) contributes to the so-called "retinal hazard region" and can cause damage to the retina, while wavelengths outside this region (i.e., ultraviolet and far infrared spectrum) are absorbed by the anterior segment of the eye causing damage to the cornea and/or to the lens. The extent of ocular damage is determined by the laser irradiance, exposure duration, and beam size. As laser retinal burns may be painless and the damaging beam sometimes invisible, maximal care should be taken to provide protection for all persons in the laser suite including the patient, laser operator, assistants, and observers.

Protective eyewear in the form of goggle, glasses, and shields provides the principal means to ensure against ocular injury, and must be worn at all times during laser operation. Laser safety eyewear (LSE) is designed to reduce the amount of incident light of specific wavelength(s) to safe levels, while transmitting sufficient light for good vision. In accordance with the ANSI Z136.3 (1988) guidelines, each laser requires a specific type of protective eyewear, and factors that must be considered when selecting LSE include: laser wavelength and peak irradiance, optical density (OD), visual transmittance, field of view, effects on color vision, absence of irreversible bleaching of the filter, comfort, and impact resistance. Ignorance of any of these factors may result in serious eye injury. As LSE often look alike in style and color, it is important to specifically check both the wavelength and OD imprinted on all LSE prior to laser use, especially

in multi-wavelength facilities where more than one laser may be located in the same room. Color coding of laser handpieces and LSE may help to minimize confusion. LSE should not move between laser rooms, nor should they be carried in lab coat pockets between use. The integrity of LSE must be inspected regularly since small cracks or loose fitting filters may transmit laser light directly to the eye. With the enormous

expansion of laser use in medicine, industry and research, every facility must formulate and adhere to specific safety policies that appropriately address eye protection.

Return to Contents

What are the effects of laser energy on the eye?

http://www.dermatology.org/laser/eyesafety.html (2 of 7)8/9/2004 10:33:58 AM

Laser Safety and the Eye

The site of damage depends on the wavelength of the incident or reflected laser beam:

● Laser light in the visible to near infrared spectrum (i.e., 400 - 1400 nm) can cause damage to the retina resulting in scotoma (blind spot in the fovea). This wave band is also know as the "retinal hazard region".

● Laser light in the ultraviolet (290 - 400 nm) or far infrared (1400 - 10,600 nm) spectrum can cause damage to the cornea and/or to the lens.

Return to Contents

Are there any specific symptoms of laser eye injuries?

● Exposure to the invisible carbon dioxide laser beam (10,600 nm) can be detected by a burning pain at the site of exposure on the cornea or sclera.

● Exposure to a visible laser beam can be detected by a bright color flash of the emitted wavelength and an after-image of its complementary color (e.g., a green 532 nm laser light would produce a green flash followed by a red after-image).

● When the retina is affected, there may be difficulty in detecting blue or green colors secondary to cone damage, and pigmentation of the retina may be detected.

● Exposure to the Q-switched Nd:YAG laser beam (1064 nm) is especially hazardous and may initially go undetected because the beam is invisible and the retina lacks pain sensory nerves. Photoacoustic retinal damage may be associated with an audible "pop" at the time of exposure. Visual disorientation due to retinal damage may not be apparent to the operator until considerable thermal damage has occurred.

Return to Contents

What types of laser safety eyewear are available?

Goggles:● fit tightly on the face● typically worn over vision-correcting prescription eye glasses● usually constructed with frame vents to minimize lens fogging● larger, heavier than spectacles or wraps

Spectacles:

http://www.dermatology.org/laser/eyesafety.html (3 of 7)8/9/2004 10:33:58 AM

Laser Safety and the Eye

● a frame that usually has two separate lenses with side shields● can be made with vision-correcting prescription eye glasses

Wraps:

● a frame with a single lens that covers both eyes● usually lighter than spectacles/goggles

Return to Contents

What are the technical considerations for eye safety?

There are two important concepts:

1. Maximum permissible exposure (MPE), is the level of laser radiation to which a person may be exposed without hazardous effects or biological changes in the eye. MPE levels are determined as a function of laser wavelength, exposure time and pulse repetition. The MPE is usually expressed either in terms of radiant exposure in J/cm2 or as irradiance in W/cm2 for a given wavelength and exposure duration.

● Exposure to laser energy above the MPE can result in tissue damage.● The ANSI 136.1 standard defines MPE levels for specific laser wavelengths and exposure

durations. Generally, the longer the wavelength, the higher the MPE; the longer the exposure time, the lower the MPE.

2. The Nominal Hazard Zone (NHZ) is the physical space in which direct, reflected or scattered laser radiation exceeds the MPE. LSE must be worn within the NHZ.

● In practical terms, when using dermatologic lasers the entire laser procedure room should be considered to be within the NHZ because the laser fiber or handpiece can be directed anywhere in the room.

http://www.dermatology.org/laser/eyesafety.html (4 of 7)8/9/2004 10:33:58 AM

Laser Safety and the Eye

Return to Contents

What factors should be considered when selecting specific eyewear?

1. Laser wavelength at which protection is afforded.2. Optical density (OD) of the LSE for the wavelength being used. OD refers to the ability of

a material to reduce laser energy of a specific wavelength to a safe level below the MPE. It can be expressed by the following formula:

OD = log10

(Ei /E

t)

Ei = incident beam irradiance (W/cm2) for a "worse case exposure" Et = transmitted beam irradiance (MPE limit in W/cm2)

Example: OD of 4.0 allows 1/10,000 of the laser light energy to be transmitted.

The required OD for any given laser can be determined by:

(a) calculation, (b) consulting nomograms or tables (e.g., ANSI 136.1 guidelines), or (c) consulting the laser manufacturer.

The OD of the LSE will decrease if the LSE is damaged. The damage threshold refers to the maximum protection that the LSE will provide for at least 5 - 10 seconds following noticeable melting or flame.

1. Comfort of the design to enhance compliance.2. Field of view provided by the design of the eyewear.3. Absence of irreversible bleaching when the LSE filter is exposed to high peak irradiance.4. Effect on color vision: the colored filter material may reduce color vision and contrast,

creating additional hazards. For example, certain LSE may interfere with visualizing monitoring equipment or detecting cyanosis during general anesthesia.

5. Impact resistance. LSE must be resistant to dust, heat, etc., so that they will not loose their effectiveness.

Return to Contents

Practical Pearls in Laser Eye Safety

1. Laser warning signs must be placed at the entrance to laser operating rooms.

http://www.dermatology.org/laser/eyesafety.html (5 of 7)8/9/2004 10:33:58 AM

Laser Safety and the Eye

2. Access to the laser operating room should only be granted to those individuals who have been appropriately educated in laser safety. Each laser facility must develop its own Safety Procedures to be enforced by an appropriately trained Laser Safety Officer for the facility. Safety procedures should be in accordance with ANSI and OSHA guidelines (and others, where appropriate).

3. As LSE often looks alike in style and color, it is mandatory to check the wavelength and optical density imprinted on each pair of LSE prior to its use.

4. Color coding of the laser handpiece and LSE may help to minimize confusion especially in facilities where multiple laser wavelengths are available.

5. LSE should not move between laser rooms, nor should they be carried in lab coat pockets between use.

6. LSE can be very expensive, so proper care and handling is mandatory. The integrity of the LSE must beinspected regularly since small cracks or loose fitting filters may permit the laser beam to reach the eye directly.7. The patient's eyes must always be protected from laser energy. If the patient is awake, appropriate opaque "mini"

goggles must be worn. Great care must be taken to avoid accidentally exposing the straps of the patient goggles to laser light, since this can ignite them.

8. Whenever laser energy is used in the immediate vicinity of the eye (e.g. treating eyelids) a stainless steel or lead eye shield should be positioned on the surface of the orbit after the application of a topical ophthalmic local anesthetic. Plastic patient eye shields cannot be expected to withstand the thermal and mechanical effects of

pulsed lasers, and should never be used.

Return to Contents

References

1. Anonymous. Laser energy and its dangers to eyes. Health Devices 1993; 22:159-204.2. ANSI 136.1 (American National Standards Institute ) American National Standard for the

safe use of lasers. Laser Institute of America, Orlando FL, 1986.3. ANSI 136.3 (American National Standards Institute ) The safe use of lasers in health care

facilities. Laser Institute of America, Orlando FL, 1988. [1995 revision will be released soon]

4. OSHA Instruction PUB 8-1.7. Guidelines for laser safety and hazard assessment. Occupational Safety and Health Administration. United States Department of Labor, Washington DC, 1991.

5. Sliney DH. Laser safety. Lasers in Surg Med 1995; 16:215-225.

http://www.dermatology.org/laser/eyesafety.html (6 of 7)8/9/2004 10:33:58 AM

Laser Safety and the Eye

Acknowledgements

We thank the Massachusetts General Hospital Dermatology Laser Center for providing us with some of the photographs used in this poster.

Back to the Top

This poster reproduction is hosted by DermWeb at UBCWe invite your comments & suggestions.

Created 10Feb96Revised: 24Dec96

http://www.dermatology.org/laser/eyesafety.html (7 of 7)8/9/2004 10:33:58 AM

Diode-Pum

pedSolid-State Lasers

Air-Cooled Ion Lasers

Helium

Cadmium

LasersH

elium N

eon LasersD

iode Laser Assem

bliesD

iode Laser Drivers and

TECControllers

1 40.7A S K A B O U T O U R C U S T O M C A P A B I L I T I E SO E M Helium Neon Lasers

SPECIFICATIONS:HIGH-POWER HELIUM NEON LASER SYSTEMS

Wavelength : 632.8 nmTransverse Mode : TEM00Pointing Stability : <0.03 mrad after 30 minutesLong-Term Drift : 82% per hourNoise (rms) : <1%Noise Frequency : 30 Hz to 10 MHzInput Voltage : 100 Vac,115 Vac, or 230 Vac810% (specify)Input Frequency : 50–60 HzShock : 25 G for 11 msecOperating Temperature : 420°C to =40°CNonoperating Temperature : 440°C to =80°COperating Humidity : 0% to 90% non condensingNonoperating Humidity : 0% to 100%CDRH Class : IIIbIEC Class : 3BCE Compliance : Compliant (230-Vac version only)

$ Output power to 35 mW

$ M2 < 1.1

$ >95% TEM00 mode

$ Complete systems, including power supply

$ CE compliant (230-Vac version only)

The Melles Griot family of high-power helium neon lasers exhibitthe long life, excellent stability, and field-proven performances thathave made Melles Griot the undisputed worldwide leader in heliumneon laser technology.

The 25 LHR/P 925, 828, and 928 lasers have guaranteed per-formances of 17, 25, and 35 mW, respectively, making them idealfor Raman spectroscopy, holography, fast scanning, and test andmeasurement applications. Most are available either in randomlyor linearly polarized (with >500:1 extinction ratio) configurations,and they can be mounted in a variety of orientations without sac-rificing power or performance. Systems are convection cooled andoperate from standard 100 Vac, 115 Vac, or 230 Vac outlets.

High-PowerHelium NeonLaser Systems

Beam Beam Maximum

cw Output Diameter Divergence Mode Longitudinal

Power (1/e2) (1/e2) Sweeping Mode Spacing Noise Package PRODUCT

(mW) (mm) (mrad) (%) Polarization (MHz) (30 Hz–10 MHz) Configuration NUMBER

17.0 0.96 0.84 ≤1 Random 257 <0.5% rm Cylindrical 25 LHR 925

17.0 0.96 0.84 ≤1 Linear 257 <0.5% rms Cylindrical 25 LHP 925

25.0 1.23 0.66 5 Linear 165 <1.0% rms Rectangular 25 LHP 828

35.0 1.23 0.66 5 Linear 165 <1.0% rms Rectangular 25 LHP 928

High-Power Helium Neon Laser Systems

*Add the appropriate suffix to the product number to indicate input voltage: -461 for 100 Vac, -249 for 115 Vac, -230 for 230 Vac.

w w w . m e l l e s g r i o t . c o m

*

40Ch_HeNe_f_v3.qxd 6/8/2005 11:21 AM Page 40.7

Dio

de-P

umpe

dSo

lid-S

tate

Las

ers

Air-

Cool

ed Io

n La

sers

Hel

ium

Cad

miu

m L

aser

sH

eliu

m N

eon

Lase

rsD

iode

Las

er A

ssem

blie

sD

iode

Las

er D

rive

rs a

ndTE

CCo

ntro

llers

40.8 1 A S K A B O U T O U R C U S T O M C A P A B I L I T I E SO E MHelium Neon Lasers

dimensions in mm

mounting areaindicated by

shading

4 holes 4-40!6.4 deep on 36 circle

cable length1.8 m

plane of polarization of linearly polarized lasers

10

25.4 25.4

10

637.3

f44.5

SHUTTER

OPEN CLOSED

25 LHR/P 925 high-power helium neon laser head

dimensions in mm

offon

54

241

161

SIDE VIEW

laser powerindicator

keyed on/offswitch

FRONT VIEW

remote interlock

High-power helium neon laser power supply

dimensions in mm

SHUTTER

+

_

OUTPUT

OPEN CLOSED

COOLING VENTS

HV CONNECTOR

78.74 1029.97

beam output 4 holes 4-40!6.4 on 36 circle

75.57

42.54

25 LHP 828 and 928 high-power helium neon laser head

40Ch_HeNe_f_v3.qxd 6/8/2005 11:21 AM Page 40.8

Registration Form for Class 3b and 4 Lasers

Complete this registration form for each laser

Manufacturer

Melles Griot

Model No.

25-LHP-928-249

Serial No.

9541EQ-1

Hazard Class(3b,4)

3b

Year Manufactured

August, 2006

Type (CW or Pulsed)

CW

Lasing Medium

Helium-Neon

Maximum Output

35 mW

Operational Wavelength(s) [nm]

632.8

Pulse width/repetition rate

NA

Beam Divergence

0.66 mrad

Emergent Beam Diameter

1.23 mm

Active/Inactive

Active

Purpose

Source for Quasi-Elastic Light Scattering instrument

List non beam hazards: Laser-generated air contaminants, compressed gasses, dyes/solvents, electrical, collateral & plasma radiation, noise, fire explosions, robotics, ergonomics

NA

Signoff: original on file signed by Robert Pelton Date: February 20, 2007 Each laser requires an SOPS to accompany the unit for operation and maintenance

Department: Chemical Engineering Building: JHE Room: B122 PI/Supervisor: Robert Pelton

AMERICAN NATIONAL STANDARD 2136.1-1993

Table 4

Simplified Method for Selecting Laser Eye Protection for Intrabeam Viewing (Wavelengths Between 0.4 and 1.4 pm). f

Q-Switched Lasers Non-Q-Switched Lasers Continuous-Wave Lasers Continuous-Wave Lasers Long-Term Staring Attenuation

(less than 1 hr)

Beam Attenuation OD

t Use of this table may result in optical densities (OD) greater than necessary. See 4.6.2 for other wavelengths.

* Not recommended as a control procedure at these levels. These levels of output power could damage or desfroy the attenuating material used in the eye protection. The skin also needs protection at these levels.

EN207 Wavelength OD %VLT Wavelength Mode Rating Per Pair190-579 nm 5+ 10% $120.00

KRY Red

EN207 Wavelength OD %VLT Wavelength Mode Rating Per Pair576-600 nm 5+ 15% 180-315 D L6 $140.00

DYE yes 585-595 nm 6.5+ Purple 180-400 R L4>315-400 D L4576-600 DI L4582-598 I L6

EN207 Wavelength OD %VLT Wavelength Mode Rating Per Pair190-380 nm 5+ 18% $120.00

KRR yes 606-694 nm 5+ Blue

EN207 Wavelength OD %VLT Wavelength Mode Rating Per Pair190-400nm 5+ 12% 180-315 D L7 $140.00

DI4 yes 625-850nm 4+ Blue 180-315 R L3662-835nm 5+ >315-400 D L4633nm 5+ >315-400 R L6

625-830 DR L4830-850 DIR L3625-670 I L4800-830 I L4670-800 I L5

EN207 Wavelength OD %VLT Wavelength Mode Rating Per Pair190-400nm 5+ 56% $150.00

RB3 pend 680-710nm 5+ Teal pending685-705nm 6+

KRY

0

1

2

3

4

5

6

7

200 300 400 500 600 700 800 900 1000 1100

nm

OD

DYE

0

1

2

3

4

5

6

7

200 300 400 500 600 700 800 900 1000 1100

nm

OD

KRR

0

1

2

3

4

5

6

7

8

200 300 400 500 600 700 800 900 1000 1100

nm

OD

DI4

0

1

2

3

4

5

6

7

8

9

200 300 400 500 600 700 800 900 1000 1100

nm

OD

RB3

0

1

2

3

4

5

6

7

8

200 300 400 500 600 700 800 900 1000 1100

wavelength

OD

www.noirlaser.com * www.lasereyeprotection.com * 800.521.9746 * 734.769.5565 * fax 734.769.1708 Page 2

NoIR P.O. Box 159 South Lyon Michigan 48178

Phone: 1.800.521.9746 734.769.5565 Fax:734.769.1708 www.noirlaser.com

Product Specification LaserShield DI4

Customer: Varied Product Designation (OPN): DI4 Luminous Transmittance: 11% Date of Revision: 2/12/06 Edited by: David W. Bothner

Graphs represent nominal filter characteristics

L-Rating 180-315 D L7 + R L3

>315-400 D L4 + R L6 DR 625-830 L4 DIR 830-850 L3

I 625-670 / 800-830 L4 I 670-800 L5

Optical Density 190-390nm 5+ 625-850nm 4+

662-835nm 5+ 633nm 5+

DI4

11% VLT

0

1

2

3

4

5

6

7

8

9

200 300 400 500 600 700 800 900 1000 1100

Wavelength (nm)

OD

#35 Wrap-Around #39 Large Fitover #33 Universal

#31 Small Fitover #700 Universal #36 Adjustable Universal

DIN - CERTCO

EC TYPE-EXAMINATION CERTIFICATE

Applicant:

Manufacturer's code: Model: Type of product: Test specifications:

Test number: Materhl: Configuration: Prescrtptlon lens: Total thickness: Curvature: Composition:

Front side: Intermediate layer: Inner ride:

NolR Laser Company 6 1 55 Pontiac Trail SOUTH WON, MI 48178 USA NOlR Dl4 Laser Eye Protactor, Filter DIN EN 207 Annex II of the Dirlective 8918861EEC

77650-PTB-03 and 1 1431 1 -PZA-05 Pol y m r

nein 2,l mm

Laser Type * Spectral range In nm Scale number Marking

DR * 180 to315nrn * L7 + L3 180-315D L7 + RL3NOlR CE DR >316 to 400 nm L4 + L6 >316-400 D L4 + R LB NO1R CE DR 825 to 830 nrn L4 625-830 DR L4 NOlR CE 1 625 to 870, >800 to 830 nrn * L4 625-670,>800-830 1 L4 NOlR CE 1 * >670 to 800 nm * L5 >670-800 1 L5 NOlR CE OIR * >a30 to 850 nm L3 * >830-850 DIR t 3 NOlR CE DIR * >850 to 860 nm L2 >850-860 DIR L2 NOlR CE 01 10600 nm * L2 * 10800 01 L2 FtOlR CE

We herewith cwtify that the above-mentioned model fulfills the basic requirements for health and protection laid down in the Directive of the European Communky on Personal Protective Equipment 8916881EEC.

DIN CERTCO Eye Protection and Personal Protective Equipment

06.10.05 - Olk

Kentek Diode, Ruby OverSpec

Home Sign In Your Account Customer Service Search KENTEK

Rapid Eyewear Finder

nm

0 Products $0.00

Laser Safety Laser Components Laser Accessories

Home

Browse KENTEK Products

● Laser Safety

❍ Eyewear

■ Polymeric

■ Broad Band Filters

■ Alignment Filters

■ Narrow Band Filters

● Laser Components

● Laser Accessories

Medical

Laser Helpers™

Laser Links

Laser Safety > Eyewear > Polymeric > Narrow Band Filters

You may also like...

Diode/Ruby 6102 Raven™ Spectacle

$189.00

Diode/Ruby SoftGog$189.00

Diode/Ruby OverSpecProduct #:KOS-6102 Laser Protective Eyewear for Diode, Ruby laser applications. KOS spectacle (OverSpec) fits over most Rx glasses, and features side filters and adjustable length temples. Meets ANSI Z136.1 and Z87.

● Laser: Diode, Ruby

● OD @ wavelength (nm): 5+ @

190 - 380

5+ @

606 - 694

● VLT (%): 18

● Lens Color: Blue

● Lens Type: Polycarbonate

● Style: OverTheGlass

●

Your Price: $189.00

Quantity

Home Your Shopping Cart Your Account Customer Service Privacy Policy

About Kentek Menu Name

© 2005 The Kentek Corporation. All rights reserved.

http://www.kenteklaserstore.com/diode-ruby-overspec-kos-6102.aspx2/20/2007 6:52:30 PM

Section 2

Instrument Operation

DYNAMIC LIGHT SCATTERING OPERATING INSTRUCTIONS Todd Hoare – September 2004 Revised by Yuguo Cui – December 2006 WARNING: DO NOT ADJUST THE KNOBS ON THE PINHOLE BETWEEN THE LASER AND THE SAMPLE CHAMBER OR ON THE PHOTOMULTIPLIER TUBE – ADJUSTMENTS WILL DE-ALIGN THE LASER! PART A – START-UP

1) Ensure the laser shutter (on laser) is CLOSED and the photomultiplier tube wavelength selector wheel is CLOSED (in the C position)

2) Turn the key control switch on the laser and wait for 30 min for the laser to warm up.

3) Wipe the sample vial with a Kimwipe to remove dust and insert into sample chamber.

4) Turn on index matching fluid pump for ~30s to remove any dust from the fluid. If air

bubbles are observed in the lines, add more decalin (by pipette) to the sample chamber.

5) Turn the temperature bath ON. Set the desired temperature by pressing SET/ENTER, typing in the temperature (including the first decimal point), and re-pressing SET/ENTER.

6) Turn on the power supply for the photomultiplier tube (located just on the photomultiplier

tube).

7) Double click on the Brookhaven Instruments folder on the desktop of the computer. Double click on the BIC Dynamic Light Scattering icon to open the program. The default windows which appear are your control window (top left), the cumulant particle size calculation window (mid left), the autocorrelation function window (bottom left), particle size distributions as calculated by exponential (top right) and CONTIN (mid right) sampling, and the count rate history window (bottom right).

8) Click on “File – Database”. Double click on your data folder so that all data you save is

directed to this folder (your previously saved files should be visible in the sample ID window when your folder is active)

PART B – SAMPLE MEASUREMENTS 9) Open up the laser by selecting 632.8 nm on the wavelength selector wheel on the

photomultiplier tube and the OPEN position on the laser shutter. LASERGARD GOGGLES MUST BE WORN STARTING AT THIS STEP THROUGHOUT THE OPERATION!

10) Return to the computer. Click on the “Dur” button in the control window (upper left).

Select the length of your experiment (typically between 1 and 3 minutes for standard particle size measurements) in the “elapsed time” box. Leave all other boxes unchecked.

11) Click on the “Params” button in the control window (upper left).

• Give your sample a name (sample ID), enter your name as the operator ID, and type additional identification information regarding your sample in the “Notes” box. • Input the test temperature (note: the actual test temperature may be different from the setpoint temperature at T<15°C or T>35°C – measure your sample temperature directly using a thermocouple in those cases).

• Select the suspension fluid (“aqueous” works for all water-based buffers). The viscosity and refractive index will be automatically inputted. If your suspension medium is not in the list or is a mixture of solvents, select “unspecified” as your liquid and manually input the appropriate viscosity and refractive index. • Input the angle – 90° for standard samples. • Select “Measured Baseline” as your normalization method. • Input the real and/or imaginary refractive indicies of your particles if you know them or keep the default values if you do not (as is typically the case). The default intensity distributions generated by the program will not be affected by the values you input here, only number and weight distributions. • If desired, select the dust filter cutoff number. This will allow you to filter out data points which are significantly different than previous data points already collected. Typical numbers which could be used range from 25 (a relatively sensitive filter) to 100 (a relatively unsensitive filter removing only the biggest spikes in the count rate history graph). WARNING: since light scattering relies on the natural variation in the count rate in order to measure particle size, use of a sensitive dust filter may artificially skew your result, so use with care! The “Use Dust Filter” box should remain unchecked (at least initially) in a standard experiment. • Keep all other boxes at their default values (wavelength = 632.8nm; self-beating laser, first calc. channel = 2)

12) Start a measurement by clicking on the green dot button in the control window. Check

the following parameters (stop the measurement by clicking on the red dot button in the control window if necessary to correct any of these parameters):

a. The average kcps (the “A CR (cur.)” value in the control window) should be

between100-250. If kcnts/s reading is too low:

i. Increase the concentration of your sample ii. Increase the filtering percentage of the laser beam. iii. Increase the slit size on the photomultiplier tube – the 100, 200, and 400

micron slit sizes may be used for DLS. Do the opposite for kcnts/s readings which are too high.

b. Delay times are properly set. The correlation data curve should look like a sigmoidal decay curve, with approximately the last 1/6 of the curve (on the right-hand side of the correlation data window, bottom left) being a flat baseline. If the baseline is too long or too short, stop the measurement (red dot in the control window) and click on the “Layout” button in the control window. Change the last delay time (increase if the baseline is too short, decrease if the baseline is too long) to a new, guessed value. Adjust the first delay such that approximately 4 decades (ie. a factor of 104) exists between the first and last delays. Accept changes by clicking “OK” and restart the measurement by clicking on the green dot in the control window. Repeat this process as necessary to achieve a proper baseline. Do not change any other parameters in the “Layout” window.

c. The count rate history (bottom right window) should look like a random residual

plot in statistics – if a baseline trend is observed (a) your temperature may not yet be stable or (b) your sample is aggregating. If spikes are observed in the curve, you have dust or aggregates in your sample and you may want to (a) remake your sample with a filtered solvent or (b) apply the dust filter to your data by clicking on the “Params” button in the control window and checking the “Use Dust Filter” box. Data points which lie outside the “acceptable” region you defined using the dust cutoff number will appear in grey in the count rate history window and will not be considered in the particle size calculations.

13) When a measurement is complete, the word “STOPPED” will appear in the top left corner

of the control window. A good measurement should have a “Base diff” of <0.1% (as given in the control window). The effective diameter (Eff Dia) of your particle is also given in the control window along with the estimated polydispersity of your sample.

14) Click on the “Summary” box in either of the particle size distribution windows to see the

mean diameter as calculated using the exponential (top left window) or non-negative least squares (CONTIN, middle left window) methods. CONTIN is appropriate for relatively monodisperse samples, while exponential may work better for multi-modal systems. If you know you have a bimodal distribution, select “ISDA – Double Exponential (Dblexp) – New Graph Window” from the top menu and click on the “Fn List” button in the new window which is opened. Select the data source (the title of your control window) to see the distribution. Use the same method to view the distributions generated by any of the statistical packages listed under the “ISDA” tab.

15) Save your data if desired by clicking on “File – Save Correlation Function”. To review this

data later, click on “File – Database”, select your folder (see step 8), and double click on the file you want to review.

16) To perform replicate measurements on the same sample, click on the “Clear” button in

the control window to clear the existing data and restart the correlator by clicking on the green dot button in the control window. Typically, 5-6 repeat measurements are done.

17) To change samples (MUST BE DONE IN THIS ORDER!):

a. Close laser shutter. b. Close photomultiplier tube wavelength selector wheel (should read “C”). c. Remove your sample and repeat steps 3 and 4 to insert a new sample. d. Repeat steps 9-15 to run your next sample.

PART C – SHUT-DOWN

18) If you are then finished with the machine: a. Close the laser shutter (on laser) and the photomultiplier tube wavelength

selector wheel (in the C position). b. Turn the key control switch off the laser. c. Turn off the power supply for the photomultiplier tube. d. Turn the temperature bath off. e. Exit the program – leave the computer on. f. Sign log book with date, operator, time of use, and comments on any problems

you had with the equipment. If you have any questions, contact Cui (e-mail cuiy3@mcmaster.ca, office JHE 138 ext. 26073; upstairs lab JHE 365 ext. 27020, main lab JHE 139 ext. 27036)

Section 3

Data Handling

Simple Intro to DLS Data Analysis (Bruce Weiner, Brookhaven Instruments)

Here are some hints for running and interpreting the BI-ISDA software. Let me know if you need more information. The source code for CONTIN has always been included on the hard disk, and we can supply the other source codes at your request. However, none are annotated, so any changes you wish to make you will have to work at it. Or, contact us and we can guide you to that part of the program you need to look at most to affect your changes. I. Cumulants The original lit. ref. is D. Koppel, J.Chem.Phys., 57(1972)4814; however, any of the books on PCS/DLS (Chu, Berne&Pecora, Schmitz, for example) show the development. In our version you must choose the baseline (PARAMS command button to select Calc. or Meas. AND, for Meas., M.BASE command button to select which measured baseline) and the number of channels to fit(Layout). We stop the fit at the first normalized point that is not positive. This, of course, depends on the baseline you have chosen. Therefore we are fitting G1 not G2. Our version is really four separate fits: linear, quadratic, cubic, quartic. The parameters of each fit appear in columns as a function of the fit order. I look for trends as a function of fit order, and I stop believing the fit when the trends reverse. This usually tells me if there are one, two, or, sometimes, three pieces of information that can be reclaimed for the fits. A true bimodal requires at least 3 degrees of freedom: one for each of the modes in size and a third representing the ratio by intensity of the amount in each mode. Thus, before I believe a bimodal fit in CONTIN, NNLS, etc., I want to see that at least the Cubic fit in Cumulants is required. With a polynomial fit one can, of course, never say what the shape of the distribution function looks like. Our Lognormal fit (a Gaussian in log space) requires two parameters: the median and the geometric standard deviation. With these two parameters we can plot any value on a Lognormal curve. To calculate these two parameters, we use the Eff. Diam. and Poly from a quadratic fit; and we assume no Mie scattering corrections (light scattering corrections) apply. Then we have two measured parameters and two unknowns, an easily solvable situation. The Lognormal fit is by intensity. The Lognormal is, by definition, a unimodal distribution. II. Double Exponential Fit There is no literature reference. This is a standard, nonlinear least squares fit to an assumed double exponential in G1. So there are four parameters: two pre-exponentials that represent the relative intensities; and two exponential decays. The decays, Gamma,

yield the assumed translational diffusion coefficients, from which we calculate the particle size. If the data is noisy, or if the measured function will not easily be constrained to fit the sum of two exponentials, or if the baseline selected is very far from accurate, then this fit may not converge. Over the years it has generally been agreed that the forced, double-exponential fit is the least useful; unless, you have a very, very good reason to believe that only two exponentials are a good model for the scattering system. III. Exponential Sampling (also known as the Pike/Ostrowski Method) The lit. ref. is N. Ostrowski, D. Sornette, P.Parker, and E.R.Pike, Optica Acta, 28 (1981) 1059. In this paper and references therein, the problem of fitting was attacked from the point of view of information theory. It was shown that a sum of exponentials, with experimental noise, and an uncertain baseline, was basically an insoluble problem. There is no analog solution; and, worse, any number of numerical solutions gave a fit that are about the same from a chi-square test point of view. All this was true even though the shape of the various fits >could be very different: a bimodal or a broad unimodal fit equally well. From the theory it was shown that there was a very limited amount of information that could be obtained, maybe only 2 or 3 parameters. This explained early failures to distinguish between fits of two or three exponentials, or to distinguish fits between different assumed functional forms. In this technique one varies the following parameters: N the number of exponentials; Omega, a parameter which relates the closest approach of the exponentials; and a mean size. See the lit. ref. for details. Using the positivity constraint, a grid search is performed to maximize N and Omega (a larger omega means a relatively higher resolution). The first cumulant is used to determine the zeroth order mean size for iteration. N equal 2 is the first order guess, and Omega equal 2 is its first order guess. As the fit proceeds you watch N, Omega, and Mean Size change. At the end, the distribution shape is determined as an envelope over a function determined by the number of exponentials. (Again, see the literature.). The fit is not just a sum of exponentials. This fit has a characteristic advantage and disadvantage: Exp. Sampling suffers less than CONTIN or NNLS from artifacts at high and low diameters; BUT, Exp. Samp. seems to broaden distributions more than the other two fits.

IV. NNLS, Non-negatively Constrained Least Squares The lit. ref. is I.Morrison, E.Grabowski, and C.Herb, Langmuir, 1 (1985) 496. Here a fixed sum of exponentials, say 50 in all, is chosen. The problem is linearized by having the user select an upper and lower limit for the particle sizes. In our program we make a guess based on the initial decay of the function for the lower limit and 100 times that for the larger limit, but the user can override that. So the problem is reduced to a linear fit of the pre-exponentials using the positivity constraint: all the pre-exponentials must be positive; otherwise, they have no physical meaning. Originally, the authors of NNLS made the spacing between the upper and lower size limits linear. Then they changed it to quadratic in recognition of the Pike work showing that limited resolution is all one can ever get out of a sum of exponentials. Disadvantage: A slightly noisy ACF will yield artifacts of a few nanometers. Repeating the measurement will show this artifacts either disappear, or shift size considerably. You can also refit the same data, manually choosing a lower size limit. If the peak at the low end follows the manually selected low size, it is probably an artifact. At the high end, a baseline error may produce an artifact in the high size end. Try fitting with the measured baseline as either the "autobaseline" in the 9000AT Windows software, or a group of 1 to 32 channels that the user can move to a flat portion of the ACF. Choose the portion closer to where the curve seems to have decayed into the noise. Advantage: When it works well, NNLS probably gives the highest resolution in particle sizing than any of the other algorithms. V. CONTIN The lit. ref. is S. Provencher, Computer Phys. Comm. 27 (1982) 213 & 229. CONTIN is only one of many, many types of regularization fits. In CONTIN the function that is minimized is the sum of two terms: the normal sum of residuals squared plus a term that is the sum of residuals cubed. There is a constant in front of this second term. It varies from zero (then CONTIN is equal to NNLS) to one (then the second term is as important as the first term. Provencher original suggested the constant should be 0.5, but other workers have preferred 0.2 or 0.3. So much depends on the "a priori" knowledge of the distribution shape. Like NNLS, the user may select an upper and lower limit or let the automatic guess prevail. In our fit we use 0.5 as the constant. That can be changed, but one must use the source codes and recompile. Disadvantages: Same as NNLS. In addition, CONTIN tends not to resolve peaks as well. Advantages: Very well known algorithm. Europeans love it. [This may be a disadvantage depending on the continent you admire most for scientific thinking.] Tends

to give smoother results. Therefore, suggested for polymer distributions. Also, the original CONTIN has a large number of variables that can be tweaked. These are all, repeat, all in the source code. If you want access, let BIC know and we can show you how to do that. WARNING: the more you play with the parameters, the better the fit, EVENTUALLY, for one set of data. But the next type of sample doesn't work as well. Result: Your tweaking satisfied what you believed to be true before you made the measurements. Bias, bias, bias is the word that comes to mind. VI. Summarizing Some Statements Above Over the years we found that different fits work well for different >situations. CONTIN penalizes multimode fits, and is best for broad, unimodal distributions of, say, polymers in solution. NNLS works well for particle size distributions, but only if they are not too broad, say no more than 5:1 in size. The Exp. Sampling tends to broaden any peaks it finds, but it does not create as many artifacts as NNLS or CONTIN. These artifacts are of two types: a peak well below any size that can be explained from the chemistry (always due to noisy ACFs); and peaks much larger than expected (usually due to baseline errors, although often mistaken for aggregate sizes). So we like it very much when at least two of these algorithms agree. Then we can place some confidence in them. YOU CAN NEVER, EVER hope to get complete distribution information from DLS. You can sometimes get more than a mean and measure of width. Be wary of bi and trimodals. Are they repeatable? Are they robust, meaning they appear with different first/last delay settings, changes in baseline, angles, etc.? Do they make sense (chemical, physical point of view)? If yes, then they are probably real, and you probably know, ROUGHLY, the position of each peak and ratio by intensity of the two peaks, roughly.

Procedures for Molecular Weight Determination through Zimm Plot (Xianhua Feng, March 2005, updated March 2007) Prior to measurement of MW, you should know the following information.

1. Polymer concentration: 0.01 to 10 mg/ml. Low concentration for high MW (e.g. above 1 mDa) and high concentration for low MW (e.g. 100 kDa).

2. Number of samples: polymer − at least 5 samples with different concentrations; Solvent for background subtraction; A calibration liquid, usually toluene. Solvent and calibration liquid should be filtered. Note that the filter you use must resist the solvent and calibration liquid. Polymer stock solution may be filtered; however, if filtration removes some polymer, do not filter your sample. Take care not to bring dust into the sample when you make polymer solution.

3. Polymer refractive index increment, dn/dc.

Operation Please refer to DLS operating instructions for start-up before measurement. Remember each time you insert a vial into the sample chamber, turn on index matching fluid pump for ~30s to remove any dust from the fluid. If air bubbles are observed in the tubing, add more decalin to the sample chamber. 1. Double click on the Brookhaven instruments folder on the desktop of the computer.

Double click on the BIC Zimm Plot Software icon to open the program.

2. Click on File − Database. Double click on your data folder so that all data you obtained later will be automatically saved in this folder. If you have not set up your folder, just create one now. Click on Exit button.

3. In the menubar of the main window, click on Hardware, Hardware Configuration. Click on Motor.

4. Set the goniometer to 90o by clicking on Set Detector Angle button.

5. Rotate the aperture wheel on the detector optics to the “C” position.

6. Click on the Experimental Parameters button in the lower part of the screen. Enter 10 seconds in the Duration/Repeat field. Click on the Dark Count Rate button.

7. When the dark count rate is measured, set the Duration/Repeat back to 1 second. Set the Number of Repeats to 10, the Dust Rejection Ratio to 1.33, the Dust Rejection Multiplier to 3, and the pinhole at 1 or 2 or 3 mm. Make sure that the actual pinhole is the same as what you have set. Note: the numbers 1, 2, and 3 on the wheel are in millimeter, whereas numbers 100, 200, and 400 are in micrometer. Select None for Polarization Analyzer and Out for Interference Filter. Click on OK button to return to the main menu.

8. Insert the highest concentration sample and set the lowest angle (30o) by clicking on Set Detector Angle button. Click on the Experimental Parameters button and then

Intensity button. Adjust the laser power and/or pinhole size to obtain maximum count rate ~ 106 cps, but keep it less than 1.5 Mcps. Click on OK button to save the setting.

9. Click on the Sample Parameters button in the lower part of the screen. Fill in the Sample Identification, Operator Identification, and Notes fields. Select a liquid from the pull down box or select Unspecified and then fill in the refractive index for the sample liquid if it is not among those listed. Keep default numbers for Ref. Index of Sample Cell (1.5), the Ref. Index of Vat Liquid (1.474), and Depolarization Ratio (0). If you know the depolarization ratio for your sample, enter your value instead of default value. If the sample liquid is water, you may correct the refractive indices by selecting Apply reflection correction, although the reflection correction is small. Fill in the value for dn/dc. Fill in 633 in the Wavelength field. Select Auto Save Results.

10. Select the calibration liquid and click on Calibrate Instrument button. The goniometer will automatically change to 90o. Insert the calibration liquid. For toluene, the calibration constant is 1 ~ 4 × 10-9 for count rate as low as 10 kcps, or 1 ~ 4 × 10-10 for count rate of 100 kcps. Click on OK to return to main menu. Note: do not change laser power and pinhole after you have calibrated the instrument since calibration constant is dependent on these parameters.

11. Click Angles in the menubar at the top of the screen. Click Edit Measurement Angles List. Click Delete All to remove the default list. If you have set angles before, click on Load List button and select the file you need. Otherwise, click on Set List Range button. Enter the first angle, last angle, and increment. Then click on OK to accept the changes. To change any angle in the list, click on the angle in the list and type over it. Usually the angles are set between 30o and 150o. Note: the difference between the last angle and the angle preceding it may not be equal to the increment entered. In that case, you may need to change angles in the list. Also notice that a minimum of 7 angles is required.

12. Click on Save List As and enter a filename up to 8 characters so that next time you can use this list by simply clicking on the Load List button and selecting it. Click on OK button to return to the main menu.

13. Click on Start button to initiate a measurement. A window on sample liquid measurement appears. Click on Yes button to initiate a solvent run at the various angles you set before. After solvent run, you will be asked for running your sample and entering the sample concentration. Once one sample is done, you will be reminded to change the sample and input new concentration.

14. When finished, click on Done button. Check if the data are saved by clicking on File/Database. If not, click File/Save As and enter a file description. Otherwise, click on Exit.

Data analysis 1. You can delete any single point by clicking on the point in the plot and then double

clicking on the blue, highlighted line in the list of angles. Double clicking again will restore the data and plotting point.

2. You can edit the concentration simply by overwriting it in the list box and then save the changes by clicking on File/Save or File/Save As.

3. You can change the plotting constant to determine which data points are good and which are bad. You can restore the original plotting constant simply by clicking on the Default button. Usually the default plotting constant is OK.

4. You can fit constant angle and constant concentration data to straight lines or to polynomials up to 5th order by choosing numbers in the list boxes labeled Fit Angles and Fit Concentrations.

5. Clicking on Calculate button to obtain molecular weight. To print the results, click on File/Print Report. To return to the main menu, click on Settings button.

6. Finally, return the goniometer to 90o. Shut down the instrument.

1. QELS OPERATIONAL CONSIDERATIONS The considerations enumerated below must be considered together: all the provisos must be met (e.g. a large residual value is not indicative of dust unless the fit error is adequately small). 1. Size Range Light scattering techniques are good for particles down to 10 nm size; the reliable upper limit is usually considered to be 0.5 micron. One micron is pushing the upper limit a bit. Going below 10 nm would require a very good signal to noise ratio (such small particles scatter very little light) and perhaps is best done with a multiple angle instrument. 1.2 Channel Width The Channel Width must be increased for larger (slower diffusing) particles to ensure that about 2 exponential decays are collected (this is an optimal value).

1.3 Fit Error Results for unimodal distributions with fit errors greater than 2 may not be reliable. If more data is collected the results may oscillate or may snap between unimodal and bimodal or trimodal. Reliable results for multimodal distributions require fit errors less than 1. 1.4 Baseline Adjust Value The contents of the long delay bins should be very close to zero (as indicated by a 0.0% baseline adjust). A larger value (necessitated by the section of the Gaussian fit algorithm that minimizes chi squared) indicates large particle contamination (e.g. dust, agglomeration) which may not be visible on the plot of PSD (depending on the minimum diameter and range values chosen). The presence of dust is also indicated by a large (e.g. greater than 10) residual. 1.5 Minimum Diameter and Range

The minimum diameter and range must be chosen so that the bins holding counts for the largest and smallest particle sizes are empty in the intensity weighted plot. Choosing the plot size brings with it the usual problems associated with histograms. High values (too many bins) can lead to apparent (but false) multimodal distributions. Too few bins will effectively act as a smoothing filter obscuring any true multimodal detail that may be present. Too many bins can result in holes in the distribution (i.e. bins containing no collected data). Probably the best solution to plot size choice is to use a large value (e.g. 45-60) and simply collect enough data to avoid any artifact holes in the distribution.

1.6 Smoothing for Unimodal versus Bimodal Choosing smoothing values can be difficult especially if one does not know whether to expect a unimodal or bimodal distribution. For a given data set a low smoothing value can yield a bimodal distribution; for the same data set a large smoothing value can yield a broad unimodal distribution. If the standard deviation is large (e.g. greater than 20%) with a small Gaussian chi squared value the distribution is probably a broad unimodal. If the standard deviation is large with a poor fit to the Gaussian model the distribution is probably bimodal. One can encounter data sets which are on the borderline between these two situations. In that case the best approach (probably the only approach other than guessing) is to collect more data to allow better estimates of standard deviation and chi squared to be made. 1.7 Weighting for Solid or Vesicular Particles In choosing between solid particle or vesicular weighting there is little difference for large particles. For small particles the difference in weighting function is significant. 1.8 Intensity Weighted Distribution Although the volume weighted distribution is usually of most interest one should always inspect the intensity weighted distribution. Not only should it be ensured that the intensity weighted distribution not contain any counts in the largest and smallest diameter bins (indicating off-scale particles) but also the distribution should be checked for low count number peaks of small diameter. These may be noise, but because of the very strong weighting given to the small particles during the conversion from intensity weighted results to volume weighted results these may end up appearing as sizable peaks in the volume weighted distribution. The solution is to collect more data and ensure that any small, low diameter peaks in the volume weighted distribution correspond to definite, well defined peaks in the intensity weighted distribution. 1.9 Optimal Count Rate We have found that the optimal count rate for some polymer particles is significantly lower than the 300 KHz suggested by the manufacturer. This value is appropriate for polystyrene but for lower refractive index polymers the target count rate should be 200 KHz. 1.10 Round versus Square Cells Disposable round cells give poorer reproducibility and broader apparent distributions than the square cells with optically parallel sides.

2 SINGLE ANGLE QELS DEFICIENCIES QELS techniques measure an intensity weighted distribution (angle dependent) which must be converted to the more physically meaningful volume weighted distribution (angle independent). Small errors in measurement of scattering intensity can lead to huge errors in the volume weighted distribution. This results from the following considerations: the conversion from intensity to volume weighting amplifies the small particles' scattering intensity greatly to compensate for their low scattering intensities compared to large particles; the intensity weighted distribution is strongly dependent on scattering angle; QELS techniques have rather low inherent resolution. In practice these deficiencies lead to the following effects: incorrect volume weighted distributions for skewed particle size distributions; lack of detection of sub-micron particles in the presence of very large (i.e. greater than one micron) particles (so much light is scattered by the very large particles that it overpowers the light scattered by the small particles; the detector effectively does not see the small particles); lack of detection of the larger particle size mode in a bimodal distribution (the smaller mode is amplified much more strongly during the conversion from intensity to volume weighting). Many of these problems can be overcome by measuring at several angles. Multiple angle QELS instruments are typically about twice the cost of fixed angle machines. 3 DISC CENTRIFUGE CONSIDERATIONS For low density polymer particles disc centrifugation seems to be able to give accurate results only for diameters greater than 200 nm. Low density particles less than 200 nm move by diffusion distances comparable to the migration due to the applied field in the disc centrifuge (the manufacturers usually claim that the minimum size limit for low density particles should be 60 nm). This confounds the results. Unfortunately, there is no easy way to determine from the nature of the raw or derived data whether the results are reliable or not when the indicated diameter is less than 200 nm. For example, for some polymers 140 nm particles may give the same instrument output as 60 nm particles (for the small particles the signal to noise ratio appears fine, the resolution appears fine, every aspect of the data appears normal). After investigating this we concluded that the only way to determine the accuracy of results indicating particle sizes below 200 nm is by consideration of several of the operational parameters in conjunction. The important parameters are: particle density; rotation speed; time required for separation; temperature change of spin fluid during separation (i.e. evaporative cooling can occur or a rise in temperature due to friction can happen). If you spin at high speeds (e.g. above 8,000 rpm) for long times (e.g. approaching one hour or more) or if your spin fluid increases temperature

more than a couple of degrees you should suspect the results (you should also be aware that the temperature indicated by some of the disc centrifuges is not the spin fluid temperature but the temperature of the compartment the rotor sits in). We feel most comfortable if we measure the particle size distribution (PSD) using at least two methods based on different measuring principles. The methods we usually apply are: a light scattering technique; a physical separation technique; electron microscopy. The results from the different techniques are seldom exactly the same (the differences become more pronounced as the PSD becomes greater). One does expect the results of the average sizes obtained from the various techniques to be close; further, one expects the shape of the distribution (i.e. breadth, modality) to be similar for the various techniques. 4. ELECTROPHORETIC MOBILITY MEASUREMENTS We previously used use a Coulter Delsa 440 electrophoretic mobility analyzer (from which the zeta potential can be calculated if desired) . This instrument has been replaced with a Brookhaven Zeta Plus.. This instrument can give approximate particle size values but the quality of the measurements are not as good as those from QELS instruments specifically meant for particle sizing. The Coulter unit was not really designed for PSD measurements. We have not used the Brookhaven ZetaPlus so we do not know whether the addition of zeta potential measurement would degrade its PSD measurement capabilities. Does this instrument give you the electrophoretic mobility data or does it give you the zeta potential (derived from the mobility data based on assumptions)? 5. CONCLUSIONS The choice between the various suppliers of particle sizing instruments (Brookhaven, Coulter, Malvern, Nicomp) frequently is made on the basis of who will be able to provide the best service. Do all of these companies have facilities nearby? The disc centrifuge is good for larger size particles but below 200 nm the results can be invalid for low density polymers. QELS instruments are good from 10 to 500 nm but the single angle instruments suffer from the problems enumerated above. A QELS instrument that measures at multiple angles and also does zeta potential certainly sounds attractive provided the particle sizing has not been compromised in favour of the zeta potential measurement.