Anda di halaman 1dari 84

A COMPARATIVE EXAMNATION OF THE SAFETY PROGRAMS AT UCLA,

UMN, AND UVM IN RESPONSE TO RECENT CHEMISTRY LABORATORY


INCIDENTS

A Thesis Presented

by

Victoria Carhart

to

The Faculty of the Graduate College

of

The University of Vermont

In Partial Fulfillment of the Requirements


for the Degree of Master of Science
Specializing in Chemistry

October, 2015

Defense Date: August 28, 2015


Thesis Examination Committee:

Rory Waterman, Ph.D., Advisor


Paul Bierman, Ph.D., Chairperson
Christopher Landry, Ph.D.
Cynthia J. Forehand, Ph.D., Dean of the Graduate College
ProQuest Number: 1601534

All rights reserved

INFORMATION TO ALL USERS


The quality of this reproduction is dependent upon the quality of the copy submitted.

In the unlikely event that the author did not send a complete manuscript
and there are missing pages, these will be noted. Also, if material had to be removed,
a note will indicate the deletion.

ProQuest 1601534

Published by ProQuest LLC (2015). Copyright of the Dissertation is held by the Author.

All rights reserved.


This work is protected against unauthorized copying under Title 17, United States Code
Microform Edition © ProQuest LLC.

ProQuest LLC.
789 East Eisenhower Parkway
P.O. Box 1346
Ann Arbor, MI 48106 - 1346
ABSTRACT

Laboratory safety has recently become more of an imperative in research


laboratories than it has ever been in the past. Recent accidents at several universities
have escalated the awareness of safety concerns in laboratory workspaces among the
general public and created a greater need for a stronger culture of safety in chemistry
research overall. Historically, results and publications have been the top priority of most
researchers, not laboratory safety.

This thesis discusses a number of laboratory accidents. The first happened in


December of 2008 at the University of California, Los Angeles (UCLA) and resulted in
the death of a graduate student researcher. Many safety concerns and violations
contributed to the fatality. The second accident happened in June of 2014 at the
University of Minnesota (UMN). This incident involved an explosion in a fume hood
that caused injuries to the researcher as well as a great deal of damage to the hood and
experimental setup. Various minor incidents at the University of Vermont (UVM) are
also discussed with regards to the effects on laboratory safety at the university.

Universities around the country have been able to learn from these accidents in
order to prevent similar occurrences in the future. These accidents and their safety
ramifications at UCLA, UMN, and UVM are the focus of this thesis. The safety
programs at each of these universities are examined and compared with respect to how
the incidents have facilitated necessary changes. Finally, future goals and opportunities
for the safety program at UVM are suggested.
ACKNOWLEDGEMENTS

I would first and foremost like to thank my advisor, Rory (RorRor)

Waterman, for all of his help and support throughout this process. I would not have

completed my degree if it weren’t for him. I don’t know why he put up with me for

as long as he did, but I am grateful.

I’d like to thank my friends and coworkers in RMS for continually bringing

me tea and supporting my graduate student time at the office (including cordoning

off my desk). Thank you to Francis and Shari for helping me to gather information I

used throughout this process.

Thank you to my fellow graduate students from the chemistry department

with whom I either started grad school or worked next to in the labs.

Thank you to my friends for either making dinner for when I got home late

in the evening, coming to Vermont to see me when I didn’t have time to leave, or

generally just being supportive.

Finally, I’d like to thank family, especially my parents and my brother, for

their continued love and support. When I may have doubted myself, they never

doubted me.

ii
TABLE OF CONTENTS

Page

ACKNOWLEDGEMENTS ................................................................................................ ii  

LIST OF TABLES ............................................................................................................. vi  

LIST OF FIGURES .......................................................................................................... vii  

ABBREVIATIONS ............................................................................................................ x  

CHAPTER 1: INTRODUCTION ....................................................................................... 1  

1.1. A Brief History of Safety in the Literature ............................................................ 1  

1.2. Thesis Goals .......................................................................................................... 4  

1.3. References ............................................................................................................. 4  

CHAPTER 2: University of California, Los Angeles......................................................... 7  

2.1. Incident Description .............................................................................................. 7  

2.2. Ramifications from the Incident .......................................................................... 11  


2.2.1. Legal Ramifications for the University .......................................................... 11  
2.2.2. Legal Ramifications for the Laboratory Supervisor ....................................... 12  
2.2.3. Safety Program Ramifications ........................................................................ 14  

2.3. Primary Conclusions Regarding UCLA .............................................................. 21  

2.4. References ........................................................................................................... 22  

CHAPTER 3: University of Minnesota ............................................................................ 24  


iii
3.1. Incident Description ............................................................................................ 24  
3.1.1. Explanation of Events ..................................................................................... 24  
3.1.2. Adjustments to the Experiment and Possible Root Causes ............................ 25  

3.2. Safety Ramifications Due to the Incident ............................................................ 27  

3.3. Analysis of and Changes to Preexisting Protocols .............................................. 29  

3.4. Joint Safety Team ................................................................................................ 33  

3.5. Primary Conclusions Regarding UMN................................................................ 36  

3.6. References ........................................................................................................... 37  

CHAPTER 4: University of Vermont ............................................................................... 39  

4.1. Timeline of Change in the Safety Program at UVM ........................................... 39  


4.1.1. Development of the Hazardous Chemicals of Concern Online Inventory ..... 39  
4.1.2. Development of the Chemistry Safety Committee ......................................... 44  
4.1.3. Other Changes to UVM’s Safety Program ..................................................... 47  

4.2. Further Comparisons Among UCLA, UMN, and UVM ..................................... 51  


4.2.1. Documentation/Signage for Current Experiments.......................................... 51  
4.2.2. Laboratory Inspection Procedures .................................................................. 53  
4.2.3. Greater Laboratory Training Requirements.................................................... 54  

4.3. Past Incidents ....................................................................................................... 56  

4.4. Primary Conclusions Regarding UVM................................................................ 60  

4.5. References ........................................................................................................... 60  

CHAPTER 5: CONCLUSIONS AND NEXT STEPS ..................................................... 62  

iv
5.1. Training Recommendations ................................................................................. 62  

5.2. Incident Follow Up Procedures and Reporting ................................................... 63  

5.3. Final Thoughts ..................................................................................................... 65  

5.4. References ........................................................................................................... 66  

CHAPTER 6: COMPREHENSIVE BIBLIOGRAPHY ................................................... 67  

v
LIST OF TABLES

Table 1: Ten initiatives of the CARE campaign at UMN. ............................................... 34

vi
LIST OF FIGURES

Figure 1: TOP reaction: Synthesis of vinyl lithium from the reaction of vinyl bromide and
tBuLi. BOTTOM reaction: Targeted synthesis of 4-hydroxy-4-vinyldecane from vinyl
lithium and 4-decanone. ...................................................................................................... 7

Figure 2: The fume hood that was used by the researcher at the time of the incident. ....... 9

Figure 3: The burn marks on the floor from the fire after the 2008 UCLA laboratory
death. ................................................................................................................................. 10

Figure 4: The syringe and 1.5” needle used for the transfer of tBuLi on the day of the
UCLA accident. ................................................................................................................ 15

Figure 5: Figures from the Sigma Aldrich Technical Bulletin AL-134 describing the
proper technique for transferring pyrophoric reagents using a syringe equipped with a 1-
2’ needle. ........................................................................................................................... 16

Figure 6: Map of UCLA’s main campus with laboratory buildings denoted with red pins.
........................................................................................................................................... 17

Figure 7: Damage to the fume hood due to the explosion at UMN. ................................. 25

Figure 8: SOC “In Case of Emergency” card from UMN to be completed for any
reaction. ............................................................................................................................. 29

Figure 9: SOC “Sensitive Materials” card from UMN to be completed for any reaction
that has specific hazards when mixed with other materials. ............................................. 30

vii
Figure 10: SOC “Unattended Hazards” card from UMN to be completed for any reaction
left unattended. .................................................................................................................. 31

Figure 11: SOC “Intended Conditions” card from UMN to be completed for any reaction.
........................................................................................................................................... 32

Figure 12: Recent timeline of change of aspects of the safety program at UVM. ............ 40

Figure 13: Fire damage to the fume hood from the 2007 fire at UVM. ........................... 41

Figure 14: The screen Burlington Fire Department sees when logging on to the campus
wide emergency response report online. ........................................................................... 42

Figure 15: An example of laboratory-space specific data the Burlington Fire Department
can find when choosing to view a building’s hazards. ..................................................... 43

Figure 16: Contributing factors for the creation of the laboratory safety group at UVM. 48

Figure 17: UVM’s Laboratory Health and Safety Policy’s general statement and
reasoning behind its creation. ........................................................................................... 50

Figure 18: The “Unattended Operations form used at UVM. ........................................... 52

Figure 19: Safety trainings taken by UVM affiliates from 2009 through 2014................ 54

Figure 20: Employee accidents in the chemistry department at UVM from 2006 through
2015 divided by type of injury. ......................................................................................... 56

Figure 21: Student accidents in chemistry teaching laboratories at UVM from 2009
through 2015 divided by type of injury. ........................................................................... 58

viii
Figure 22: Total trainings received by UCLA affiliates from 2007 through 2012. The
blue portion represents online trainings, and the red portion represents classroom
trainings............................................................................................................................. 62

Figure 23: Reporting process for accidents and injuries at UVM. ................................... 64

ix
LIST OF ABBREVIATIONS

BFD – Burlington Fire Department


Cal/OSHA – California Occupational Health and Safety Administration
CARE – safety campaign representing compliance, awareness, resources, and education
CSB – Chemical Safety Board
DEC – Department of Environmental Conservation
HCOC – hazardous chemicals of concern
JST – Joint Safety Team
LHAT – Laboratory Hazard Assessment Tool
Me3SiN3 – azotrimethylsilane
OSHA – Occupational Health and Safety Administration
PEG – poly(ethyleneglycol)
PI – principle investigator (laboratory supervisor)
PPE – personal protective equipment
RMS – Risk Management & Safety
SDS – Safety Data Sheet(s)
SOC – Safe Operations Card(s)
SOP – Standard Operating Procedure
tBuLi – tert-butyllithium
UCLA – University of California, Los Angeles
UMN – University of Minnesota
UVM – University of Vermont

x
CHAPTER 1: INTRODUCTION

1.1. A Brief History of Safety in the Literature

Laboratory safety has become more of an imperative in research laboratories than

it ever has been.1-4 This change is partially due to recent high profile accidents at

prestigious universities that have increased the awareness of safety concerns in laboratory

workspaces to the general public.5

There have been numerous journal articles in recent years specifically regarding

laboratory safety and ways of generating a more pronounced culture of safety in

academic institutions, both in safety education6-9 and basic laboratory safety.1, 3, 10-11

However, there are only a few articles from the early 1900s that strictly deal with general

laboratory safety.12-17 These articles are outdated based on modern safety standards,

though there were various tips and ideas that are still useful in chemistry laboratories

today. For example, a discussion of how to properly store potentially hazardous

chemicals appeared in 1925.13 The article discusses detailed demonstrations to perform

for researchers to understand their laboratory hazards and how improper storage could

create a laboratory fire. One such example was to form a flat disk of dried ammonium

nitrate on an asbestos sheet, cover with a layer of dried zinc dust, put one drop of water in

a hollow in the chemicals, and wait for there to be a reaction.12 There are other ways of

describing what improper storage could lead to in a laboratory, but as Davison points out

in the article, “the thought of safety in storage will be quickened by those who try out

some of the demonstrations here recorded.”12 As safety progressed into being a higher

1
concern in laboratories, these types of demonstrations became obsolete as much as

unnecessarily risky.

As initially pointed out in the article, one portion of any safety program should

include proper chemical storage. There are various charts and tables that are available to

help determine how chemicals should be stored and with which other chemicals.18-20

Such distinctions highlight two aspects to storage, reactivity of different classes of

chemicals and regulations (including requirements such as labeling). Some charts focus

more on the reactivity of different classes of chemicals or two individual chemicals.18 For

instance, the tables on the University of California, Los Angeles (UCLA) website give

examples of what chemicals should not be stored together (e.g. cyanides and acids).18 The

University of Vermont (UVM) and the University of Minnesota (UMN) focus on the

specific storage of the chemicals but in conjunction with how to properly label the

containers.19-20 UMN in particular focuses a good deal on the regulatory reasoning for

why chemicals need to be where and how they are stored, referencing the Laboratory

Safety Plan, Employee Right to Know, and the National Fire Protection Association in

the storage guide.19

Another article from 1925 concentrated on treating students that had been

exposed to a poisonous chemical.13 There are suggested treatment options that include

counteracting the ingested poison with other chemicals. A treatment of “alkalies” is

suggested, “[giving] dilute acid, hydrochloric, acetic, vinegar, lemon juice, orange juice

or a fixed oil” among others to the exposed researcher.13 Among the options medical

2
attention is not included, which would be the principal suggestion today. Many of the

suggested treatment options directly oppose what current safety data sheets (SDS)

describe as best practices for what should, or should not, be done. For instance, there is a

disparity between the article and the current SDS for sulfuric acid. Turner’s article

suggests inducing vomiting, which is in conflict with the SDS: “Rinse mouth. Do NOT

induce vomiting.”13, 21 This article is most similar to potential trainings researchers may

receive regarding what to do in an emergency. In emergency response training at the

University of Vermont, various emergencies are discussed, including what to do in case

of exposure, fire, injury, etc.a Turner’s article, though the information is outdated, was in

an attempt to make researchers more proactive and prepared for any actuality.

Key lessons to take away from these early articles related to laboratory safety are

that there is always a need for emergency preparedness and there is always room for

improvement, especially as new information becomes available. At the time the articles

were published, the suggestions and information would likely have been the best

available. Rather than using prior practices because they are established, awareness

about current laboratory safety developments and the hazards involved needs to be

increased to mitigate risk.

a
The list of topics discussed in emergency response trainings is from UVM’s safety
training, “Emergency Response for Laboratory Workers.”
3
1.2. Thesis Goals

The analysis of recent serious accidents at the University of California, Los

Angeles (UCLA) and the University of Minnesota (UMN) is the initial focus of this

thesis. The safety programs at these two universities and the University of Vermont

(UVM) will be compared and contrasted as well as accidents at all three universities.

Once the similarities and differences have been discussed, future goals and opportunities

will be suggested for UVM.

The goal of this analysis is to determine gaps in the safety program at UVM with

regards to the program modifications made over the last five years at UVM, UCLA, and

UMN. The safety gaps and areas for improvement are to be informed by the incidents at

the three universities.

1.3. References

1. Cooper, M. D.; Phillips, R. A., Exploratory Analysis of the Safety Climate and
Safety Behabior. J. Saf. Res. 2004, 35 (5), 497-512.

2. Hill Jr., R. H., The Emergence of Laboratory Safety. J. Chem. Health Saf. 2007, 14
(3), 14-19.

3. McGarry, K. A.; et. al. Student Involvement in Improving the Culture of Safety in
Academic Laboratories. J. Chem. Educ. 2013, 90 (11), 1414-1417.

4. Mulcahy, M. B.; et. al. College and University Sector Response to the U.S.
Chemical Safety Board Texas Tech Incident Report and UCLA Laboratory Fatality.
J. Chem. Health Saf. 2013, 20 (2), 6-13.

4
5. Christensen, K., Dealdy UCLA Lab Fire Leaves Haunting Questions. Los Angeles
Times 2009.

6. Hill Jr., R. H., Getting Safety into the Chemistry Curriculum. Chem. Health Saf.
2003, 10 (2), 7-9.

7. Hill Jr., R. H., Strengthening Safety Education of Chemistry Undergraduates.


Chem. Health Saf. 2005, 12 (6), 19-23.

8. Nelson, D. A., Incorporating Chemical Health and Safety Topics into Chemistry
Curricula. Chem. Health Saf. 1999, 6 (5), 43-48.

9. Banister, M. R., An Environmental, Health and Safety Class for Undergraduate


Chemistry Majors. Chem. Health Saf. 2005, 12 (2), 20-22.

10. DeJoy, D. M.; et. al. Making Work Safer: Testing a Model of Social Exchange and
Safety Management. J. Saf. Res. 2010, 41 (2), 163-171.

11. DeJoy, D. M.; et. al. Creating Safer Workplaces: Assessing the Determinants and
Role of Safety Climate. J. Saf. Res. 2004, 35 (1), 81-90.

12. Davison, H. F., Safety First in Storing Chemicals. J. Chem. Educ. 1925, 2, 782-783.

13. Turner, R. G., Suggestions to the Teacher of Chemistry on Poisons and Their
Treatment. J. Chem. Educ. 1925, 2, 466-471.

14. Walker, F., Safety in Chemical Laboratories. J. Chem. Educ. 1934, 11, 506-509.

15. Buxbaum, E. C., Safety in the Chemical Laboratory. J. Chem. Educ. 1934, 11 (2),
73-76.

16. Keller, E., The Hygiene of the Small Chemical Laboratory. Industrial &
Engineering Chemistry 1910, 2 (6), 246-251.

17. Turrill, P. L., Public Liability and Chemical Education. J. Chem. Educ. 1933, 10
(9), 552-555.
18. UCLA Chemical Safety Storage Guidelines.
https://http://www.ehs.ucla.edu/research/lab/chem/storage.

19. UMN The Safe Chemical Storage Guide.

5
20. UVM Segregate and Store Chemicals Safely. http://www.uvm.edu/safety/lab/label-
and-store - segregate%20&%20Store.

21. Sigma-Aldrich Safety Data Sheet: Sulfuric acid.


https://http://www.sigmaaldrich.com/MSDS/MSDS/DisplayMSDSPage.do?country
=US&language=en&productNumber=320501&brand=SIAL&PageToGoToURL=h
ttps%3A%2F%2Fwww.sigmaaldrich.com%2Fcatalog%2Fsearch%3Fterm%3Dsulf
uric%2Bacid%26interface%3DAll%26N%3D0%26mode%3Dmatch
partialmax%26lang%3Den%26region%3DUS%26focus%3Dproduct.

6
CHAPTER 2: University of California, Los Angeles

2.1. Incident Description

On December 29, 2008, during the holiday break, a deadly accident occurred at

UCLA that was preventable. The researcher involved in the accident was scaling up a

reaction she had performed that October after starting work in Professor Harran’s

laboratory.1 The reaction, according to her notebook, is shown in Figure 1. The initial

step of the reaction was to generate vinyl lithium via the reaction of vinyl bromide with

two equivalents of tert-butyl lithium (tBuLi).1

Figure 1: TOP reaction: Synthesis of vinyl lithium from the reaction of vinyl bromide and tBuLi.
BOTTOM reaction: Targeted synthesis of 4-hydroxy-4-vinyldecane from vinyl lithium and 4-
decanone.

As mentioned, when the reaction had been completed in October, it was on a

smaller scale. In that run, anhydrous ether (28 mL) was added to a dry 200 mL round

bottom flask. Vinyl bromide (3.0 mL) was added to the anhydrous ether in the flask and

stirred at –78 °C. After 15 minutes, 53.79 mL of 1.67 M tBuLi in pentane was added to

the mixture and stirred for another 2 hours, followed by increasing the temperature of the

solution to 0 °C for 30 minutes, and finally returning the solution to –78 °C. In a separate

7
flask, ether (6 mL) and 4-decanone (3.90 mL) were mixed, cooled, and transferred to the

flask containing the solution of vinyl lithium. The reaction was stirred at –78 °C for two

hours. The temperature was then increased to –10 °C before quenching the excess tBuLi

with sodium bicarbonate. This reaction yielded 3.60 g of 4-hydroxy-4-vinyldecane at

86.75% crude yield.1 In December, the objective was to increase the reaction scale to

produce an overall yield of about three times as much product.

The reaction in December required the researcher to use 159.5 mL of 1.69 M

tBuLi in pentane. The total volume was obtained using a 60 mL syringe to measure

approximately 50 mL at a time in three additional steps. The researcher was working in a

fume hood (Figure 2) with a Schlenk line. Also located in the hood, there was an open

flask of hexane.

8
Figure 2: The fume hood that was used by the researcher at the time of the incident.

The initial problem began when the plunger came out of the syringe and the tBuLi, a

known pyrophoric reagent at the concentration. The solution ignited when it came into

contact with air. When this happened, the flask of hexane was spilled and the solvent

ignited. Unfortunately, the researcher was wearing neither a flame-resistant laboratory

coat nor other protective clothing, and her sweater caught fire. Though there was a safety

shower in the laboratory, it was not used. Instead, another researcher in the laboratory

tried to smother the flames on the researcher with his laboratory coat. When this did not

appear to be working, he poured water on her from a nearby laboratory sink.2

9
Throughout the incident, the researcher was reported to be conscious and even

speaking at times to the other members of her group.1 When the emergency responders

arrived on the scene, the fire was extinguished though the damage was already done (see

Figure 3 for the damage to the laboratory alone).

Figure 3: The burn marks on the floor from the fire after the 2008 UCLA laboratory death.

The medical personnel used the safety shower for decontamination of the researcher, then

she was brought to UCLA Ronald Reagan Medical Center, and finally, she was

transferred to Grossman Burn Center — over 40% of her body was covered in 3rd degree

10
burns. She died about two-and-a-half weeks later, on January 16, 2009, due to the

injuries she suffered on that day.1

2.2. Ramifications from the Incident

There were many legal and non-legal implications for both the university and the

laboratory supervisor resulting from this incident. However, aside from the legal details,

the potential dangers of working in a chemistry laboratory were thrust into the public’s

view. This story made headlines nationwide in chemistry journals and among chemists,

but also in mainstream news outlets.3 This incident was something that could not be

ignored and for good reason.

2.2.1. Legal Ramifications for the University

In 2009, the California Occupational Health and Safety Administration

(Cal/OSHA) determined that the accident was a result of safety lapses and inadequate

training.2, 4
The researcher’s training should have minimally included how to properly

handle pyrophoric reagents including tBuLi as well as where to find the emergency

equipment and how to use it if necessary (e.g., the safety shower). For the university, this

resulted in the required creation of a $500,000 scholarship in the researcher’s name and a

complete overhaul of the safety program (Section 2.2.3).2 Cal/OSHA’s investigation cited

poor training, poor technique/improper method, lack of supervision, employees not

wearing required personal protective equipment (PPE), and flammable liquids and

volatile chemicals stored improperly.1, 5 The lengthy list of citations highlighted the need

11
for many of the changes that were consequently made to the safety program at UCLA as

a direct result of the investigation.

Aside from the costs of the scholarship and safety program adjustments, the

university was required to pay legal fees. Though there was no direct punitive action

against UCLA, the legal fees alone were about $4.5 million covering both the university

and the principle investigator (PI) for the laboratory.6-7 In comparison to what the

university has spent on laboratory safety increases, updates, and required changes, the

legal fees were small. The laboratory safety adjustments have cost the university

upwards of $20 million dollars as of October 2014.7 The use of this money, in updates

consistent with current best practices, will be outlined subsequently in Section 2.2.3.

2.2.2. Legal Ramifications for the Laboratory Supervisor

Separate from the penalties against the university—including the scholarship and

changes to the safety program—the laboratory supervisor, or principle investigator (PI),

was charged with four felony counts of willful violation of state occupational health and

safety standards in the December of 2011, the first time in which a PI faced criminal

charges for failure to meet standards.6 As a result, this incident was reported in

mainstream media coverage.3 Many of Cal/OSHA’s citations, mentioned above, were

debated throughout the legal case with the PI. For instance, as laboratory supervisor, the

professor is responsible for any training that is required for researchers working in the

laboratory.8 The supervisor does not necessarily need to deliver the trainings personally,

but he or she must ensure that required trainings have been completed before work

12
commences. The case against Professor Harran was settled after the better part of five

years in June of 2014. Two punitive results of the settlement were 800 hours of

community service and a $10,000 fine to be paid to the burn unit where the researcher

was treated.8 Other outcomes from the trial require Professor Harran to teach a

preparatory organic chemistry course to inner city high school graduates and speak about

laboratory safety to new graduate students in the chemistry department at UCLA.8

When researching this accident and its subsequent outcomes, it was evident that a

dialogue had been started involving both chemists and non-chemists. The community

responded to not only the outcome but also the progression of the case as it was

happening. Some responses to the case’s outcome were that the researcher should be

responsible for him/herself and know the reagents and hazards that will be handled before

running any experiment.8 However, most responses questioned if the penalties were

enough of a punishment for what happened, explaining that the PI should have been more

aware of the safety concerns in his laboratory and trained his researchers accordingly.9-13

This case has had such a large impact on the safety at UCLA as well as the safety culture

around the country. For this reason, some commented that the penalties against Harran

were not severe enough.9-13 Legally, the laboratory supervisor is ultimately responsible

for what happens in his laboratory.6 Thus, if someone is injured due to improper

technique or training, ignorance regarding the use of safety equipment, or a similar lack

of knowledge regarding laboratory hazards, the researcher’s current supervisor is

13
responsible, regardless of any training a given researcher may have received at another

institution.

The legal side of the argument also varied greatly. Some contended that Professor

Harran should not have been initially charged so severely.10, 12 Others debated whether

the outcome of this case would be a cause for change at any universities outside of

California.10, 12-13
Since this was the first time a PI was criminally charged due to a

laboratory accident, there are likely repercussions that will not be known until future

accidents occur at universities.

2.2.3. Safety Program Ramifications

Because Cal/OSHA found that the accident was a result of safety lapses and

inadequate training, the safety program and culture at UCLA were examined very

thoroughly by the school and Cal/OSHA, among other interested parties.3 It was

determined that the proper protocol was not being followed at the time of the incident.

According to the laboratory supervisor, the laboratory workers follow Sigma-Aldrich

“Technical Bulletin AL-134” regarding the safe handling of air-sensitive reagents,

including tBuLi.14 According to the protocol, glassware used with pyrophoric materials is

dried in an oven rather than flame-dried, which was the method used according to the

researcher’s notebook. Either method should remove any water adsorbed to the surface

of the glassware; however, flame-drying was not part of the accepted Sigma-Aldrich

protocol the laboratory had adopted. For small quantity transfers (up to 50 mL), a

14
syringe is acceptable to use for the transfer but only when using a 1–2 foot long needle.

The needle that was utilized was 1.5 inches long and can be seen in Figure 4.1, 14

Figure 4: The syringe and 1.5” needle used for the transfer of tBuLi on the day of the UCLA
accident.

A needle this short would require a different method than what is described in the

technical bulletin of obtaining the tBuLi from the primary container. A longer needle

would be able to reach the bottom of the bottle easily, thereby avoiding the need to invert

the bottle, which could damage the septum or create a risk of dropping the flask (see

Figure 5).

15
Figure 5: Figures from the Sigma Aldrich Technical Bulletin AL-134 describing the proper technique
for transferring pyrophoric reagents using a syringe equipped with a 1-2’ needle.

Through no more than 50 mL was transferred at any given time, a cannula transfer would

have been the more accepted method to transfer the total amount required, 159.5 mL.

Using a cannula would have allowed the researcher to maintain an inert atmosphere while

handling the pyrophoric materials as little as possible. Exposure can more easily be

avoided with the use of a cannula.14

The specifics mentioned above regarding changes to and/or enforcement of the

protocol are minor changes in the program as compared to the broader transformation

UCLA’s safety program experienced. One month following the accident, inspections

were performed in over 300 laboratories at the university, many in the chemistry

16
department. At the top of the list of operational problems was the lack of documentation

concerning how many total laboratories and researchers were on campus.15 In a large

campus with laboratories spread out among multiple buildings (Figure 6), this is not

necessarily surprising, though it is a serious problem.

Figure 6: Map of UCLA’s main campus with laboratory buildings denoted with red pins.16

A Laboratory Safety Committee was created, and within six months of the

incident, five ambitious recommendations were laid out for the university. These five

recommendations involved 1) the development of a safety culture at the university that


17
would generate safer laboratories, 2) the increase of training and outreach, 3) the increase

of accountability and oversight of the safety staff, 4) the change to make safety a top

priority in laboratory design, and 5) the increase of the accuracy and amount of

inventories and recordkeeping in laboratories on campus.15, 17-18


Once these were

enumerated, UCLA needed to overcome a major hurdle: there was a large disconnect and

lack of trust between the safety department/staff and the laboratory researchers. A

partnership between safety staff and laboratory researchers was emphasized as the

ultimate goal of this portion of safety culture at UCLA.15

One major change that increased the partnership between safety staff and

laboratory researchers was the implementation of a standardized inspection procedure.15

Regardless of what kinds of hazards are present in the laboratory being inspected

(chemical, biological, radioactive, etc.), the inspection questions and, ultimately, the

report are either the same or comparable. When there is less freedom to personalize

inspections, there is greater accountability from each inspector. Uniform inspection

documents ensure systematic expectations for researchers and their laboratories,

regardless if the same inspector performs each inspection every year. Additionally,

senior laboratory safety officers within the safety department at UCLA double-check the

laboratory inspections and make random selections to re-inspect various laboratories.

This duplication by separate safety officers guarantees reproducibility of the results and

determines where there are gaps to be filled.

18
After the investigations by Cal/OSHA and UCLA, the Chemical Safety Board

(CSB) also weighed in on the incident. The CSB released a video entitled

“Experimenting with Danger” that discussed three chemistry laboratory incidents: the

fatal incident at UCLA that has been discussed in this thesis, a fatal accidental poisoning

at Dartmouth College in 1997, and an explosion at Texas Tech University that severely

injured a graduate student in January 2010.5, 19-20 This video focused on why the incidents

occurred, e.g. lack of training, improper use of personal protective equipment, necessary

changes to procedure when scaling up a reaction, etc. More than why the incidents

occurred, the video focused a great deal on the need for a strong safety culture. The CSB

emphasized that by implementing and supporting a strong culture of safety, preventing

hazards and exposures becomes a higher priority. Because universities often do not fall

within the general limitations set by OSHA and other regulatory bodies, universities are

forced to develop their own allowances and limitations.

The CSB has determined six key safety lessons to take from incidents like those

discussed in their video, “Experimenting with Danger:”

1. An academic institution … should ensure all safety hazards, including


physical hazards of chemicals, are addressed.
2. Academic institutions should ensure that practices and procedures are
in place to verify that research-specific hazards are evaluated and
mitigated.
3. Comprehensive guidance on managing the hazard unique to laboratory
chemical research in the academic environment is lacking. …
4. Research-specific written protocols and training are necessary to
manage laboratory research risk.
5. An academic institution’s organizational structure should ensure that
the safety inspector/auditor of research laboratories directly report to an

19
identified individual/office with organizational authority to implement
safety improvements.
6. Near-misses and previous incidents provide opportunities for education
and improvement only if they are documented, tracked, and
communicated to drive safety change.5, 20  

Lesson 4, “researcher-specific protocols and training are necessary to manage laboratory

research risk,” has been demonstrated at UCLA in their actions since the 2008 accident.20

The CSB’s intent is for the research being conducted in the laboratory to be examined to

determine how to prevent exposure. Safety Data Sheets can be useful in this practice but

do not give enough of the necessary information regarding how to properly and safely

handle a hazardous chemical. The work practices required in a laboratory need to be

explicitly described for all researchers.

After the accident in 2008, UCLA was required by Cal/OSHA to spend time and

money to further develop their safety culture.15 Part of this process has been to develop

more standard operating procedures (SOPs). One use of SOPs is to describe in detail the

proper work practices that a researcher should use in the laboratory to maintain safety.

Currently, there are hundreds of chemicals included in UCLA’ SOP template library.21 At

the top of each template, there is a disclaimer: “This is an SOP template and is not

complete until: 1) lab specific information is entered into the box below 2) lab specific

protocol/procedure is added to the protocol/procedure section and 3) SOP has been

signed and dated by the PI and relevant lab personnel.”21-22 These stipulations make clear

that each SOP needs to be particular to the laboratory that will be using it, and that all

personnel must sign their acknowledgement of required use. By signing an SOP, a

20
researcher is acknowledging required use (of said SOP) but also acknowledging all of the

potential hazards of the chemical in question.

SOPs can be used to assist in training about the hazards of the chemical and what

to do in case of emergency. UCLA updated their chemical hygiene plan in 2014: “[The

plan] establishes a formal written program for protecting laboratory personnel against

adverse health and safety hazards associated with exposure to potentially hazardous

chemicals…”23 The many SOPs included in the template library will help to promote the

protection of laboratory workers. However, no two laboratories will be able to use the

same SOP for a given chemical because the set up, procedure, and other reagents or

solvents being used with it may differ.

2.3. Primary Conclusions Regarding UCLA

After the fatal accident in December of 2008, UCLA was obliged to make

numerous changes to its safety program as discussed in this chapter. One of the most

useful changes was to better enforce the use of SOPs and the following of proper

procedures and protocols. SOP enforcement should make researchers more accountable

for what reactions and procedures are performed in the laboratory. Since these

documents can be used as training tools, it allows for multiple researchers to get the same

information and learn the same techniques regardless of when the training occurs. This

should necessarily increase the consistency in the laboratory when performing the same

procedures. In the future, the university will need to continue to increase the

21
documentation of trainings. This accident has promoted UCLA to create a stronger

culture of safety. Other universities around the country have also been able to use this

unfortunate accident as a learning tool.

2.4. References

1. Kemsley, J. N., Learning from UCLA. Chem. Eng. News 2009, 87 (31), 29-31, 33-
34.

2. Morris, J. Landmark Worker Death Case Continues Against UCLA Chemistry


Professor. http://www.publicintegrity.org/2012/07/27/10313/landmark-worker-
death-case-continues-against-ucla-chemistry-professor.

3. Christensen, K., Dealdy UCLA Lab Fire Leaves Haunting Questions. Los Angeles
Times 2009.

4. Cal/OSHA Citation and Notification of Penalty.


http://pubs.acs.org/cen/_img/87/i31/Cal-OSHA_UCLA_citation_2009-05-04.pdf.

5. USCSB, Experimenting with Danger. 2011.

6. Kemsley, J., Lab Death Legal Defense Cost University of California Nearly $4.5
Million. Chem. Eng. News 2014, 92 (44), 9.

7. Trager, R., UCLA Spent $4.5 Million on Legal Costs in Sangji Case. Chemistry
World 2014.

8. Trager, R. UCLA Chemist Avoids Prison Time for Lethal Lab Accident.
http://www.rsc.org/chemistryworld/2014/06/chemist-avoids-prison-sheri-sangji-lab-
safety-case-settlement.

9. Bracher, P., UCLA Professor Patrick Harran Charged in Sheri Sangji's Death. In
ChemBark, 2011.

10. Bracher, P., UCLA Professor Patrick Harran Strikes Deal with Prosecutors. In
ChemBark, 2014.

22
11. Chemjobber, ChemBark: Not Enough Punishment for Harran. In Chemjobber,
2014.

12. Blum, D. Bad Chemistry: an Update on the Sheri Sangji Case Wired [Online], 2013.
http://www.wired.com/2013/07/a-bad-chemistry-update/.

13. Bracher, P., UCLA Chemistry Professor Patrick Harran to Stand Trial. In
ChemBark, 2013.

14. Sigma-Aldrich, Technical Bulletin AL-134: Handling Air-Sensitive Reagents.


2012.

15. Gibson, J. H.; Schroder, I.; Wayne, N. L., A Research University's Rapid Response
to a Fatal Chemistry Accident: Safety Changes and Outcomes. J. Chem. Health Saf.
2014, 21 (4), 18-26.

16. UCLA General Services Campus Map. http://maps.ucla.edu/campus/.

17. UCLA UCLA Laboratory Safety Committee.


https://http://www.ehs.ucla.edu/research/lab/committee.

18. UCLA Report to the Chancellor of UCLA Laboratory Safety.


https://ucla.app.box.com/ehs-report-to-chancellor.

19. USCSB "Experimenting with Danger" Focuses on CSB Case Study on Texas Tech
University Accident; Laboratory Deaths at UCLA and Darmouth.
http://www.csb.gov/csb-releases-new-video-on-laboratory-safety-at-academic-
institutions/.

20. USCSB Texas Tech University Laboratory Explosion.


http://www.csb.gov/assets/1/19/csb_study_ttu_.pdf.

21. UCLA Standard Operating Procedure (SOP) Template Library.


http://www.sop.ehs.ucla.edu.

22. UCLA, Standar Operating Procedure: tert-Butyllithium. 2012.

23. UCLA UCLA Chemical Hygiend Plan. https://ucla.app.box.com/ehs-chemical-


hygiene-plan.

23
CHAPTER 3: University of Minnesota

3.1. Incident Description

3.1.1. Explanation of Events

The incident, a laboratory explosion with personal injury and property damage,

at the University of Minnesota (UMN) occurred on June 17, 2014. The reaction was

started on June 16th and was intended to generate approximately 200 g of

azotrimethylsilane (Me3SiN3) in poly(ethyleneglycol) (PEG) solution.

Chlorotrimethylsilane was reacted with sodium azide to produce Me3SiN3 with sodium

chloride as the byproduct (Equation 1).1-2

(1)

The reactants were combined in a distillation apparatus and were stirred overnight. On

the 17th, the solution was heated slowly with stirring using a magnetic stir bar. The

researcher performing the reaction was walking from his office, through the laboratory, to

the hallway and noticed that the thermometer was crooked. He went to adjust the

apparatus, and the reaction exploded. The explosion caused injuries to the researcher,

including second-degree burns, injuries to one arm and side from broken glass, and an

injured eardrum from the blast. There was also damage to the building, including the

destroyed hood and experimental setup (see Figure 7).2

24
Figure 7: Damage to the fume hood due to the explosion at UMN.3

3.1.2. Adjustments to the Experiment and Possible Root Causes

The researcher had performed this reaction on numerous occasions, and while

there is some speculation regarding the cause of the explosion,2, 4 specific changes to the

reaction will be discussed first. When this reaction was originally performed, the scale

was smaller, where about half as much chlorotrimethylsilane and sodium azide were

used.1-2 Similar to the incident at UCLA, the researcher did not make adjustments to the

procedure due to the scaling up of the reaction. In both the UCLA and UMN cases, the

resulting injuries and property damage would have been somewhat mitigated by a smaller

25
amount of hazardous chemicals in use since there would have been less of the hazards

present.

Another adjustment the UMN researcher made was to the diethylene glycol

dimethyl ether that was originally used as the solvent. Diethylene glycol dimethyl ether

was used in the literature and in prior runs;5 however, in an effort to give better mixing,

the solvent was changed to PEG with a molecular weight of about 300 g/mol. This

solvent change was also reported in the literature but only when producing about half the

desired product – 100 g instead of 200 g.6 When multiple variables in a reaction are

changed at the same time, there is a greater risk of something going wrong.

One possible explanation for the blast is that hydrazoic acid may have been

generated in the reaction vessel. Hydrazoic acid (HN3) is shock sensitive, and upon

heating, explosive.7-8 Though it was not one of the reagents added to the flask, it can be

generated when sodium azide reacts with acid. The solvent may have reacted with

sodium azide to generate hydrazoic acid in solution, which was being heated.2, 7 Another

similar explanation is that the PEG might not have been completely anhydrous, despite

the fact that a newly opened bottle was used. If this was the case, the water could have

provided the requisite protons needed to produce hydrazoic acid in the heated solution.2

An alternative explanation was that the magnetic stir bar had stopped moving,

which would allow unreacted sodium azide to settle on the bottom of the flask and be

overheated.2 Heated sodium azide can violently release nitrogen, which may have caused

the explosion. This possibility is realistic due to the researcher noticing the crooked

26
thermometer. If the stir bar had been working correctly, it is reasonable to theorize that

the thermometer should have remained in its correct position. If the solution had been

continuously stirred, unreacted sodium azide would not have been able to settle to the

bottom of the flask.

3.2. Safety Ramifications Due to the Incident

This incident happened about five and a half years after the death at UCLA. The

ramifications for UCLA had been an impetus for other universities around the country to

increase their cultures of safety to prevent similar incidents from occurring in their

schools’ laboratories. To this end, UMN created the Joint Safety Team (JST) (see

Section 3.4) in April of 2012, two years before the explosion.9 Similar to UCLA, UMN

administrators and safety officials implemented immediate changes to its safety program

to help prevent reoccurrences of the UCLA incident or similar accidents. The change

that may yield the most impact is the required use of “Safe Operations Cards” (SOC).

The cards are “used as a tool to complete hazard assessment and planning for research

activities.”10 These cards were originally developed by Dow Chemical in an effort to

stimulate safe practices in research laboratories. The SOCs were made available to UMN

due to a partnership with the chemical company. Dow Chemical Company formed

eleven safety partnerships in 2011 with universities across the country. This partnership

between Dow and UMN has UMN receiving $17 million from Dow over a five-year

period to “strengthen research and enable a building expansion.”11 UMN researchers had

27
sparingly used these cards before, but the incident has served as a catalyst for wide spread

adoption due to increased enforcement.4 SOCs need to be completed for any reactions

completed in the laboratory. There are blank sections to be completed by the researchers:

reaction, contact information, intended conditions, hazards, and emergency shutdown.

This card was based on Dow’s creation and has been adjusted and edited to be as useful

as possible for all laboratories at UMN (see Section 3.3).

Since this incident, Professor William Tolman, the chair of the UMN chemistry

department, has initiated a limit on the scale of this particular type of reaction, and any

similar reactions, to no larger than 5 g theoretical yield.1, 4 Scaling up of reactions is quite

common in research laboratories, especially for the process of synthesizing starting

material. Issues arise when the unexpected reactions/outcomes materialize at the larger

scale. When using relatively small amounts, any potential damages from unintended

reactions can be more easily managed. Because the reaction was on a large scale, the

explosion that occurred had more material available that was able to react violently. The

greater amount of material ultimately resulted in scope of the injuries and damages to the

researcher and the laboratory, respectively. Limiting the scale of potentially dangerous

(e.g. explosive) experiments will prevent large-scale unintended reactions from

happening again, thereby reducing the overall risk of personal injury and property

damage.

28
3.3. Analysis of and Changes to Preexisting Protocols

Aside from the SOC mentioned above, there are four other “cards” than can, and

should, be posted on laboratory doors or fume hoods to give any non-researchers, such as

emergency responders, or other researchers, such as those not performing the experiment

in question, information about what is happening in the laboratory. “In Case of

Emergency,” “Sensitive Materials, Avoid,” Unattended Hazards,” and “Intended

Conditions” are the four card options.

On the “In Case of Emergency” card (Figure 8), there is contact information, but

there is also a section to describe an emergency shut down procedure.

Figure 8: SOC “In Case of Emergency” card from UMN to be completed for any reaction.10
29
A number of options are listed that can be checked off (e.g. “evacuate lab,” “drop lab

jack,” “turn off ___,” etc.). The “In Case of Emergency” card allows anyone not familiar

with the particular reaction the ability to keep everyone safe while also maintaining the

integrity of the reaction.

The “Sensitive Materials” card gives options to be checked, similar to the

emergency card, but in this case, it is for what types of materials to avoid (Figure 9).

Figure 9: SOC “Sensitive Materials” card from UMN to be completed for any reaction that has
specific hazards when mixed with other materials.10

30
This card allows anyone outside the laboratory to understand what should not be

introduced to the laboratory space with respect to whatever is currently being done or

materials stored in the laboratory. For instance one option that can be checked is “Air,”

meaning that whatever is in the laboratory should not come into contact with atmospheric

air. If the sensitive reaction were in a hood, this card would make it clear that it must be

kept under an inert atmosphere to avoid potentially dangerous reactions.

For any reaction left unattended, the “Unattended Hazards” card will describe

what the potential hazards are (Figure 10).

Figure 10: SOC “Unattended Hazards” card from UMN to be completed for any reaction left
unattended.10

31
This card could be especially useful for emergency responders. If, for instance, the fire

alarm in the building is tripped, fire fighters will be able to determine if the reactions

occurring in the laboratory could be the reason for the alarm or if they should move on to

the next laboratory.

The “Intended Conditions” card can be used to describe the intentional

conditions of the reaction (Figure 11). For instance, there is an option of “Keep

DARK/covered” that would help anyone to avoid destroying work by exposing the

reaction to light. This card can also be used to describe what the reaction should look

like in terms of both condition (stirring) and appearance (color).

Figure 11: SOC “Intended Conditions” card from UMN to be completed for any reaction.10

32
If one of the intended conditions were not met another researcher would be able to notify

the person performing the experiment that something incorrect is happening. The

experimenter would then be able to assess the situation to determine if the laboratory is

safe to occupy to continue work. These cards can be posted on the laboratory door so

that there is no need to enter the laboratory to assess the condition of the area. By using

the same template on every laboratory door in a building, emergency responders can be

trained to look for any hazards listed on the cards before entering a laboratory, thereby

keeping everyone safer from potential exposures to hazardous materials.

3.4. Joint Safety Team

As mentioned above, two years prior to the chemical explosion, UMN had

developed the Joint Safety Team (JST).9, 12-15 The JST consists not of faculty and staff,

but rather of graduate student and postdoctoral researchers from two departments, 1)

chemistry and 2) chemical engineering and mathematical science. One objective of this

group is to improve on safety culture at UMN. Similar to the conditions noted in Chapter

2, safety culture not only refers to the manner in which experiments are executed but also

to attitudes and the perception of those inside and outside the laboratory, other

researchers, staff members, etc. Following the explosion in the fall of 2014, JST

implemented a new safety campaign called CARE, which represents four safety areas:

Compliance, Awareness, Resources, and Education.9, 16 Each area focuses on particular

33
safety-centric themes that can be overlooked in research laboratories. There are ten

initiatives that have been introduced within the CARE campaign (see Table 1).

Table 1: Ten initiatives of the CARE campaign at UMN.9

Initiative Description CARE Category


Identify 10 guidelines Document describing most important aspects
Awareness
for a safer lab of lab safety
Highly attended event to introduce the JST
Kickoff event Awareness
and identify new goals for safety culture
Templates to display hazards and contact Compliance and
Standard lab signage
information for each lab space Awareness
Brief discussion or presentation on a safety
Awareness and
Safety moments topics at the beginning of group meetings
Education
and departmental seminars
A short safety tip sent weekly by email to Awareness and
Weekly safety note
faculty, staff, and students Education
5 unique posters on safety topics designed Awareness and
Safety posters
and displayed Education
Laboratory Semiannual student-let audits to evaluate
Compliance
Walkthroughs laboratory housekeeping and compliance
Website containing links, information, and Resources and
JST website
JST content Education
Annual session to train LSOs in their roles Resources and
LSO training
and responsibilities Education
Week focused on reducing clutter and waste
Cleanup week Resources
from labs, in conjunction with DEHS

One of the ten initiatives that can help promote a strong culture of safety is the

“Weekly Safety Note,” which is a brief safety tip sent to faculty, staff, and students via

email.9 If this is sent out early in the week, it could encourage researchers to keep safety

in mind all week while planning and performing various experiments in the laboratory.
34
An analogous promotion to safety is the “Safety Moment” that is either discussed at

group meetings or during departmental seminars. In either format, the Safety Moment is

at the beginning of the meeting or seminar. Similar to the weekly safety note, the safety

moment discussed can create a stronger safety culture by increasing awareness. One

example of how to use a Safety Moment is to discuss a near miss that happened in the

laboratory. A near miss can be described as anything just short of an accident.17-18 An

example is if a researcher drops a flask containing nitric acid on the floor, but the flask

does not break. The laboratory group can learn how to prevent future accidents from the

near misses that occur. Discussing the near miss, or other safety related topics, can lead

to the researchers caring more about safety in general. By having these discussions at the

beginning of the meeting instead of the end, it can emphasize the importance of a strong

safety culture by maintaining that it is not an afterthought. Both of these initiatives are

under two CARE categories: Awareness and Education.

An interesting initiative under the category of Compliance is “Laboratory

Walkthroughs.” Students lead a biannual audit assessing housekeeping and compliance

in the laboratory.9 By having the students run the audit process, there is a greater sense of

responsibility and ownership they undertake to ensure the laboratory is in compliance

with regulations. This initiative also encourages the students to be more aware of the

changes in regulations to stay in compliance whether that awareness is through the JST’s

other initiatives or not. Similarly to UCLA’s objective of creating a greater partnership

35
among safety staff and researchers, this type of audit process will inevitably yield a

stronger partnership.

There is one initiative in particular in the last category, Resources, that likewise

encourages cooperation between the safety staff and researchers. The JST has initiated a

“Cleanup Week” that encourages laboratory groups to remove clutter and waste from

their laboratories.9 This cleanup can be used to eliminate excess hazards and decrease the

chance of an accident from occurring or escalating. Laboratories are able to remove any

hazardous materials that are no longer needed, thereby eliminating the hazard altogether.

The next chapter will in part discuss the similarities between UMN’s Cleanup Week and

UVM’s Chemistry Safety Day.

3.5. Primary Conclusions Regarding UMN

Though UMN had created the Joint Safety Team before the incident in their

chemistry department, new lessons were learned from the outcomes of the explosion.

Similar to UCLA, enforcement of using proper protocol and procedures will now be more

thoroughly applied, and new protocols and procedures have been developed to avoid

similar occurrences from happening in the future. The JST has been a strong source of

safety awareness at the university. If it continues to be a presence and keep safety a

priority, the number of incidents and accidents will likely decrease.

36
3.6. References

1. Kemsley, J. More Details on the University of Minnesota Explosion and Response.


http://cenblog.org/the-safety-zone/2014/07/more-details-on-the-university-of-
minnesota-explosion-and-response/.

2. Taton, T. A.; Partlo, W. E., Chemical Safety: Explosion Hazard in Synthesis of


Azidotrimethylsilane. Chem. Eng. News 2014, 92 (43), 2-3.

3. Kemsley, J. Explosion Injures University of Minnesota Graduate Student.


http://cenblog.org/the-safety-zone/2014/06/explosion-injures-university-of-
minnesota-graduate-student/.

4. Kemsley, J., C&EN Talks Safety with Minnesota Chemistry Chair William B.
Tolman. Chem. Eng. News 2014, 92 (45), 34.

5. Birkofer, L.; Wegner, P., Trimethylsilyl Azide. Org. Synth. 1970, 50, 107.

6. Chen, C.-Y.; et. al. Synthesis, DNA-Binding Abilities and Anticancer Activities of
Triazole-pyrrolo[2,1-c][1,4]benzodiazepines Hybrid Scaffolds. Bioorg. Meg. Chem.
Lett. 2013, 23 (24), 6854-6859.

7. Wiss, J.; et. al. Explosion and Decomposition Characteristics of Hydrazoic Acid in
the Gas Phase. Org. Process Res. Dev. 2007, 11 (6), 1096-1103.

8. Kemsley, J. Sodium Azide + Acid = Boom. http://cenblog.org/the-safety-


zone/2010/04/sodium-azide-acid-boom-2/.

9. UMN Joint Safety Team: About. http://www.jst.umn.edu/about.html.

10. Dow Safe Operation Card (SOC). http://storage.dow.com.edgesuite.net/safety-dow-


com/External-SOC-Form_v060612.pdf.

11. UMN U of M to Receive Nearly $17 Million Over the Next Five Years from Dow
Chemical Company. http://discover.umn.edu/news/business-law/u-m-receive-
nearly-17-million-over-next-five-years-dow-chemical-company.

12. Kemsley, J., Dow Chemical Teams Up With Universities On Laboratory Safety.
Chem. Eng. News 2012, 90 (44), 21-23.

37
13. UMN Joint Safety Team. http://www.jst.umn.edu.

14. UMN Graduate Students and Postdoctorates Lead Laboratory Safety Efforts.
http://cse.umn.edu/admin/comm/features/2012_9_19_dowchemical-safety.php.

15. UMN Dow + U = Lab Safety: A New Safety Campaign for Laboratories Takes
Hold. http://discover.umn.edu/news/vision-leadership/u-and-dow-chemical-team-
lab-safety.

16. UMN Safety Initiative Becoming Embedded into Department's Culture.


http://www.chem.umn.edu/news/news.lasso?serial=642.

17. UCSB Near Miss Reporting. http://www.chem.ucsb.edu/about/safety/near-miss.

18. UDel Near Miss Reporting.


http://www.udel.edu/ehs/generalhs/indhygiene/nearmiss-reporting.html.

38
CHAPTER 4: University of Vermont

4.1. Timeline of Change in the Safety Program at UVM

4.1.1. Development of the Hazardous Chemicals of Concern Online Inventory

The safety program at the University of Vermont (UVM), similar to other

universities, focuses not only on the chemistry department but all departments with

hazards including chemical, biological, radioactive, and physical. The Laboratory Health

and Safety Policy states UVM’s objective as “providing a healthy and safe working and

learning environment, and to supporting environmentally sound practices.”1 When

incidents occur at comparable universities, there are opportunities to strengthen the safety

culture locally. For instance, after the incident at UMN, a safety professional from UMN

delivered a seminar to the chemistry department at UVM to share experiences, results,

and actions. Discussing the possible causes of an incident and how to prevent

recurrences in the future allows both universities to be better prepared. Incidents like the

explosion at UMN create opportunities to create a dialogue among safety staff and

laboratories researchers.

Incidents at UVM and various universities have contributed to the evolution of the

safety program at UVM. The timeline shown in Figure 12 shows a very general and

basic analysis of the safety program at UVM between 2007 and 2012.b In that five-year

span of time, the program began to evolve into the program it is today.

b
The information in the timeline was sourced from documentation maintained at UVM’s
Environmental Safety Facility.
39
Figure 12: Recent timeline of change of aspects of the safety program at UVM.

One incident at UVM that sparked a great change in the safety program was a

laboratory fire. A fire started in a non-metal fume hood Memorial Day Weekend, 2007

and caused $2 million in damage/repairs and two years of delayed work. The smoke that

was generated by the fire was exhausted out of the building by the fume hood. Hence,

the smoke detectors outside of the fume hood in the laboratory were unable to detect the

fire until the fume hood was burning (Figure 13).

40
Figure 13: Fire damage to the fume hood from the 2007 fire at UVM.c

Fortunately, there were no deaths or serious injuries of either researchers or emergency

personnel, but several firefighters received emergency care. After the fire, UVM’s online

chemical registry system was developed. Every researcher responsible for a laboratory

must now register that space – list all researchers in the laboratory and complete the

Hazardous Chemicals of Concern (HCOC) online chemical inventory. The Burlington

Fire Department (BFD) can access this inventory on route to any UVM building to see

what hazards are present (Figure 14).

c
Photograph is from Risk Management and Safety’s archives at UVM.
41
Figure 14: The screen Burlington Fire Department sees when logging on to the campus wide
emergency response report online.2

If there was a similar system in place before the fire in 2007, the need for emergency

medical treatment could have been more easily avoided. BFD can now see what hazards

are located in the building before or during a fire alarm or emergency call response. The

responders are safer as a direct result of updated inventories and being able to contact

someone responsible for the laboratories in a building.

Once the building to which they are responding is chosen, BFD is able to view

all hazards associated with individual laboratory spaces (see Figure 15).

42
Figure 15: An example of laboratory-space specific data the Burlington Fire Department can find
when choosing to view a building’s hazards.3

All inventories are required to be updated a minimum of every six months to ensure

accuracy. The implementation of the online inventory is one part of the annual training

BFD receives from UVM Risk Management and Safety personnel. A better partnership

has been created between UVM and BFD since the fire and its resulting impacts.

Similar to UCLA’s Laboratory Hazard Assessment Tool (LHAT) the HCOC

allows for the transfer of information from researchers in a particular laboratory to a non-

laboratory member of the university. One difference between the HCOC and LHAT is

the latter focuses on hazard assessment as an informative tool as opposed to hazard

communication.4 Hazard assessment refers to the evaluation of the hazards present in the

laboratory and is accomplished by LHATs. Hazard communication refers to the


43
communication of information with anyone that may contact the hazards and is achieved

by HCOC. LHATs somewhat combine the usefulness of an SOP with informative power

though the communication of that assessment. The evaluation of the hazards is

completed through filling out the LHAT form. Therefore, there is an opportunity to add

value to the HCOC by increasing the information that is required to be made available,

including assessing the hazards present. However, there is a fine line between offering

enough information to be useful and giving too much information. At this point, UVM is

focused on keeping the information updated to ensure that what information is present is

actually useful for emergency responders because it is accurate. BFD is able to see an

abbreviated version of the HCOC so that they are able to assess the building’s hazards

before entering it.2 Forms like the LHAT have too much information to determine and

extract the vital information to use as a quick assessment tool.

4.1.2. Development of the Chemistry Safety Committee

Another development at UVM that happened prior to the accident at UCLA was

the creation of the Chemistry Safety Committee in August of 2008. Committee

membership includes seven to ten chemistry faculty/staff, the Laboratory Safety

Coordinator for the College of Arts and Sciences, and the Assistant Director of Risk

Management & Safety for Health and Safety, and it is chaired by a chemistry faculty

member. The Chemistry Safety Committee holds biannual meetings to examine various

safety topics ranging from recent incidents to upcoming events and seminars to ongoing

concerns. One objective of the Chemistry Safety Committee is to encourage safe work

44
practices in the chemistry department. In other words, the committee looks to increase

the strength of the safety culture within all chemistry laboratories.

The Chemistry Safety Committee is similar to the Joint Safety Team at UMN in a

few ways. One aspect of the Chemistry Safety Committee that is comparable to the Joint

Safety Team is that both are focused on safety culture and its improvement in chemistry

laboratories. However, where the Chemistry Safety Committee is all faculty and staff,

the Joint Safety Team is comprised of only graduate students and post-doctoral

researchers.5 The Chemistry Safety Committee could gain some insight through another

viewpoint by adding graduate students/post-docs to the committee. This different

perspective could add value to the Chemistry Safety Committee by ensuring that all

needs in the department are acknowledged.

The committee at UVM has already had some success in addressing its objective

of encouraging safer work practices. For example, the committee has implemented

Chemistry Safety Day. At this annual, daylong event, the entire chemistry department

uses that time to update required trainings, participate in various drills, attend a

departmental safety seminar, and engage in laboratory cleanup. Laboratory cleanup

consists of cleaning out the laboratory of old chemicals, wastes, and papers. The first

annual Chemistry Safety Day was held in May of 2012, and this has been an annual event

ever since. The first Safety Day supplied researchers with the ability to take live fire

extinguisher training and to remove old chemicals, wastes, papers, and broken equipment

from the laboratory. Chemistry is the only department to currently require the live fire

45
extinguisher training to be completed by all of its researchers, faculty, and teaching

assistants. The first Chemistry Safety Day in 2012 had participation from 88% of all

researchers in the fire extinguisher training.

Of the more than 500 laboratories at UVM, the chemistry department has some of

the highest potential hazards. This department has maintained its priority status, which

includes receiving yearly safety audits. Events like Chemistry Safety Day help to

minimize hazards and increase hazard awareness among undergraduate and graduate

students, postdoctoral researchers, faculty, and staff.

The committee focuses mainly on the research laboratories in chemistry, but it

explored how to increase safety awareness in undergraduate students taking chemistry

laboratories as a portion of general chemistry and organic chemistry classes as well. The

goal is to engage undergraduate students in laboratory safety as soon as (or before) the

students enter the laboratory. There was a small safety book that had been used in all

beginner undergraduate laboratories, and while there was some good information

contained, the book itself did little to encourage the students to take an interest in the

topic of safety. In an effort to better train undergraduate students in safety, the committee

developed a new online training tool. Two chemistry faculty members, two chemistry

staff members, and the laboratory safety coordinator for chemistry created an online

training that the students are now required to complete and need to achieve a minimum

score of 80% on an assessment prior to engaging in experimental work in the laboratory.

The training is in the format of a PowerPoint presentation, which is displayed via the

46
Blackboard online learning platform, where students are able to access the training and

complete the assessment. The training itself includes the basic safety information for

instructional laboratory work including, personal preparedness, how to recognize

chemical hazards, SDS information, routes of exposure, engineering and administrative

controls, proper PPE, what to do in case of a chemical spill, housekeeping, how to

properly handle chemical waste, and emergency equipment/what to do in case of an

emergency. The feedback from the students taking the new training has been very

positive, especially any students that had previously used the safety book mentioned

above.

4.1.3. Other Changes to UVM’s Safety Program

While the Chemistry Safety Committee has been beneficial to the promotion and

improvement of safety culture within UVM’s chemistry department, there are other

factors that have facilitated changes in the safety program at the university as a whole.

As noted, the accident at UCLA was one contributing factor in the creation of UVM’s

Laboratory Safety Policy but not the only factor. In 2011, a laboratory safety working

group was created specifically to further develop laboratory safety. This group was

created as a result of safety concerns at UVM as well as incidents at both UCLA and

Yale6, d (see Figure 16).

d
In 2011, a 22-year-old undergraduate was working in a chemistry laboratory machine
shop when her hair was caught in the lathe, resulting in the researcher’s death.35
47
Figure 16: Contributing factors for the creation of the laboratory safety group at UVM.

The accident at UCLA discussed in Chapter 2 made the need for a robust safety policy

quite evident. This was one of two fatal accidents at prestigious universities that

contributed to the safety policy. The accident at Yale in 2011 involved a researcher

working in a chemistry shop on a lathe when her hair was caught in the spinning

mechanism.6 The types of laboratories vary greatly at UVM, and these accidents clearly

showed a need for a stronger safety culture. The third factor was the ending of Project

XL.e Project XL, which stands for “eXcellence and Leadership,” was a national initiative

that tests innovative ways of achieving better and more cost-effective public health and

e
“The focus of [Project XL] is on the implementation of an institution-wide laboratory
environmental management program which effectively minimizes, reuses, collects and
disposes of waste chemicals from campus teaching and research activities.”36
48
environmental protection.7 Project XL determined the waste management at the

university, and it “was developed because the standard hazardous waste regulations

established under the Resource Conservation and Recovery Act were preventing

laboratories in higher education from fully developing their pollution prevention

programs.”7 There was potential for a big shift in waste management due to the impeding

end of this project.

The fourth and fifth factors were the Department of Environmental Conservation

(DEC) inspection of the UVM campus and the specific inspection of the chemistry

building found a number of non-compliance issues that needed to be handled. The issues

included missing documentation, labeling problems or old waste, self-inspections were

not completed, and two laboratories had significant housekeeping concerns. Overall, the

inspection cost UVM almost $20,000 in fines. The combination of these five factors

(Figure 16) led to the creation of the Laboratory Health and Safety Policy (Figure 17).

There were 15 members of the Laboratory Safety Group at the time of its

inception including representatives from general safety, radiation safety, chemical safety,

laboratory safety, biosafety, and environmental compliance; the Director of Risk

Management; two Assistant Deans for Research (College of Medicine and College of

Arts and Science); the Vice President for Research; and the Chief Compliance Officer.

The working group was convened to generate a robust Safety Policy. The reasoning

behind its creation is outlined on the first page of the policy as can be seen in Figure 17.1

49
Figure 17: UVM’s Laboratory Health and Safety Policy’s general statement and reasoning behind its
creation.1

In the last few years, UVM has made a number of changes and updates to its

safety program. For example, the organizational structure within the Risk Management

& Safety (RMS) Department has been adjusted to allow for safety staff members to spend

more time with individual researchers when appropriate and needed. There is a greater

opportunity for one-on-one trainings for new principle investigators (PIs) and laboratory

safety officers in the laboratory. New trainings have been created and older trainings

have been updated, both classroom and online trainings. More targeted trainings allow

for researchers to gain specific insight before working with particular hazardous materials

or equipment. Furthermore, the safety websites have almost all been updated within the

50
last five years. All of these updates and new creations have been prepared to ensure that

researchers are getting the most up-to-date information possible and that they can easily

learn about any new regulations that would affect their work or their safety procedures in

the laboratory.

4.2. Further Comparisons Among UCLA, UMN, and UVM

4.2.1. Documentation/Signage for Current Experiments

The safety program at UVM is comparable in various ways to the safety

programs at both UCLA and UMN. As previously mentioned, the HCOC at UVM is

similar to the LHAT at UCLA, though the first is more focused on hazard communication

and the latter is more about hazard assessment. Also, the Chemistry Safety Committee

and the Joint Safety Team were both created for the purpose of promoting and improving

safety culture. However, there are other comparisons to be made. For instance, UVM

uses an “Unattended Operations Form” to define what procedures are being performed in

the laboratory without any researcher present (Figure 18).

51
WARNING!
UNATTENDED OPERATIONS TAKING PLACE

INSTRUCTIONS
An Unattended Operations Sign must be filled out completely and attached to the
laboratory door facing the hallway whenever any equipment, experiment, process or
operation is left unattended and could pose a potential risk to health, safety or security of
personnel or property. Users must post all applicable information, including emergency
shut-down instructions. For more info, contact safety@uvm.edu

This information is used to inform other building occupants, Physical Plant personnel,
Safety staff and other Emergency Responders in the event of an unexpected reaction,
overheating, odor or emergency.

Date: Start Time: ________ (circle) AM PM


Name: End Time: ________ (circle) AM PM

Phone: Will Check Again: in #_____min’s,


in #_____hrs, in #_____day(s)
PI/Supervisor Name: PI/Supervisor Phone:

The Unattended Operation taking place involve the following hazards:


Circle ALL that apply and list specifics as requested.

Equipment Hazard
Muffle Furnace Temp: ___________ (circle) F or C
What exactly is being heated in oven/furnace?
______________________________________
Chemical Fume Hood Chemicals in use (circle all that apply) Amount: ________
Corrosive , Toxic, Reactive, Flammable
Compressed Gas Gas Type: _______________________
Vacuum Pump Type of oil: ______________________
Electrical Volts/Watts: __________ (circle) AC or DC
Hot plate/Ignition Sources Type/Temp: ___________________________
Continuous Water Flowing
Other: Describe “other”:

Provide an additional description of what is taking place:


______________________________________________________________________________

Emergency Shut off is located: ___________________________________________________

Questions about this form? Contact safety@uvm.edu or go to uvm.edu/safety

Figure 18: The “Unattended Operations” form used at UVM.8

52
This form is similar to the “Unattended Hazards” SOC used at UMN. The Unattended

Operations form is to be used “whenever any equipment, experiment, process or

operation is left unattended and could pose a potential risk to health, safety or security of

personnel or property.”8-9 This form was originally created in 2001, but it has been

updated as recently as 2012. Both forms are used to display information to anyone

outside the laboratory. If something happens in the laboratory, researchers, safety staff,

or emergency responders would all be able to better assess the risks of entering the

laboratory.

4.2.2. Laboratory Inspection Procedures

Another area of comparison is the safety inspection procedures at each university.

As mentioned, the UVM chemistry department laboratories are high priority. For

auditing purposes, that means safety staff inspects the laboratories every year. Beginning

in 2012, LabCliQ has been used to track the inspections and corrective actions

required/completed for all UVM laboratories online. Thus, the same questions are

answered regardless of the laboratory’s department or which safety professional is

completing the inspection. Adoption of the online platform was done to normalize the

auditing process, though the standardization can only go so far. In a similar manner,

UCLA laboratories are inspected using a standardized approach.10 The difference at

UCLA is that there are monthly reviews of recent inspections, and another member of the

safety staff redoes an inspection randomly chosen.10 This is to ensure the reproducibility

of the audit process. As at UVM and UCLA, UMN has an auditing checklist that has

53
minimal open-ended questions to be answered.11 This, again, allows for a more

normalized auditing process. Comparing the different schools’ processes, it is clear that

normalization is the goal.

4.2.3. Greater Laboratory Training Requirements

There have been increases in the number of safety trainings at all universities

discussed here. At UVM, training has increased greatly, as illustrated in Figure 19.

Safety  Training  by  Year  and  Platform  


11000  
Total  Number  of  Training  Completed  

10000  
9000   1859  
1822  
8000  
7000  
6000  
1270  
5000   Classroom  
4000   877   806   8365   7910   Online  
3000   881  
4754  
2000   3822   3736  
1000   2542  
0  
2009   2010   2011   2012   2013   2014  
Year  

Figure 19: Safety trainings taken by UVM affiliates from 2009 through 2014.

The large increase in the number of completed trainings from 2012 to 2013 was partially

due to the increased number of required trainings for all laboratory workers and better

enforcement of the requirements. This was a period of catching up with trainings for
54
many researchers. Another explanation for the large increase from 2012 to 2013 is the

implementation of the annual online refresher training. 2013 was the first year the

refresher was offered and required for all researchers that were at UVM since 2012 or

earlier. The number of completed trainings will likely drop slightly over the next few

years as researchers that have been at UVM for a number of years finish completing the

basic training requirements. Instead, new students, faculty, and staff will complete the

majority of all recorded trainings, and near the same number each year will complete the

annual refresher trainings. Similar to UVM, at UCLA and UMN trainings have been

added to the requirements for all laboratory workers. This increase in safety training

activities is targeted at ultimately generating a higher level of safety awareness.

The undergraduate training that is displayed through BlackBoard is not recorded

as part of the total trainings delivered by RMS. Roughly 3000 undergraduate chemistry

students take the safety training each academic year through BlackBoard before

performing any laboratory experiments for class. This number includes general and

organic chemistry students. This number is slightly inflated since there are students

repeating the trainings from general chemistry when they take organic chemistry.

Therefore, any student taking organic chemistry will see the same information twice,

hopefully reinforcing the culture of safety that is being cultivated.

55
4.3. Past Incidents

UVM safety staff utilizes and learns from incidents that occur at other

universities, however, there have been a number of incidents at UVM and within the

chemistry department specifically. The incidents are reported based on the injured

person’s status – either employee or student. For employees, there have been 16

incidents between 2006 and 2012 (Figure 20).f

Past  Incidents:  Employees  in  Chemistry  


6  

5  
Number  of  Incidents  

4   BBP  Exposure  
Hernia  
3  
Sprain/Strain  
2   Needle  Stick  
Chemical  Exposure  
1  
Laceration  
0  
2006   2007   2008   2009   2010   2011   2012   2013   2014   2015  
Year  

Figure 20: Employee accidents in the chemistry department at UVM from 2006 through 2015 divided
by type of injury.

f
The data for Figure 20 was compiled through RMS documented reports of the incidents.
56
There are not enough data points to determine any significant trends. There were no

reported incidents in 2013, 2014, and through June 2015. Lacerations were the most

common injury, 7 of the 16 accidents. The number of incidents from year to year appears

to be random, fluctuating from one to five without any steady increase or decrease in

occurrence. Continued analysis of these reports could potentially yield a trend in the

future.

Though undergraduate student accidents have not been the focus up to this point

in the discussion, UVM follows up with various injuries and accidents in the teaching

laboratories as necessary. RMS’s “follow up” process endeavors to determine the root

causes of accidents. This causation determination may be through meeting with the

researcher, observing performance of the task or experiment that was done during the

accident, or sampling for exposure. There have been 59 reported incidents in teaching

laboratories from April of 2009 to April of 2015 (Figure 21).g

g
The data for Figure 21 was compiled through RMS documented reports of the incidents.
57
Past  Incidents  in  Chemistry  Teaching  
Labs  
14  

12  
Number  of  Incidents  

10  
Fainting  
8  
Burn  
6  
Needle  Stick  
4   Chemical  Exposure  

2   Laceration  

0  
2009   2010   2011   2012   2013   2014   2015  
Year  

Figure 21: Student accidents in chemistry teaching laboratories at UVM from 2009 through 2015
divided by type of injury.

Similar to the employee records, lacerations were the most common injury, 22 of the 59

incidents. Though the number is more than a third of the total, most of the students did

not require emergency attention and care. Chemical exposures were the second most

common injury with 20 incidents being a result of exposure. Again, the high variability

and small sample size makes it very difficult to determine any significant trends. There is

a slight decrease in the number of total incidents from year to year; however, this could

be for a variety of reasons. A possible explanation is that the follow up procedures done

by both RMS and the chemistry department are having a positive effect on the awareness

58
of safety in the laboratory. The chemistry department in particular has made an effort to

remove more hazardous chemicals/experiments from the teaching curriculum. By

eliminating some of the hazards, the department is able to mitigate injuries from

occurring. Another possibility is that the injuries are simply not being reported. Injuries

that do not require medical attention are less likely to be reported since there is no need

of a formal report for insurance purposes. If injuries go unreported, the number of

incidents would be deceptively low.

In cataloging this data, an opportunity for UVM to enhance the follow up

procedure that is used at the university was identified. For instance, there is no single

form or procedure that is followed by all safety personnel when performing a follow up.

Though the questions asked may be similar and the same information may be gleaned

from the process, a more uniform procedure would be best. There is another opportunity

in the way the incidents are reported. To date, the incidents are difficult to find and trend,

regardless of the size of the data set. A more user-friendly reporting format for both

employee and student reports is needed. It is insufficient to be simply reporting the

accidents if reports cannot be found easily and aggregate data obtained. If all incidents

were collected in the same database, potential trends could be found. This analysis,

absent from current procedures, would ultimately help to prevent serious injuries and

accidents from recurring in the future.

59
4.4. Primary Conclusions Regarding UVM

UVM has been motivated by the incidents that have happened at the university

and other universities around the country to further develop the safety program. Various

accidents are discussed in RMS-delivered classroom trainings to make researchers aware

of what is progressing around them at similar universities. It is emphasized that

researchers need to be more aware in general of their surroundings, especially when

working with hazardous chemicals, materials, or equipment in the laboratories on

campus. UVM uses not only the accidents that happen in chemistry laboratories or with

hazardous chemicals but other hazards as well to ensure all students, faculty, and staff are

working safely with the hazardous materials or equipment.

However, data collection is insufficient to determine trends and learn what types

of accidents can and should be prevented. It is not enough to collect the data, it also

needs to be evaluated and cataloged in a useful way. If UVM is to continue to learn from

accidents, there needs to be a better way to evaluate steps that need to be done to avoid

reoccurrences.

4.5. References

1. UVM Laboratory Health and Safety.


http://www.uvm.edu/policies/riskmgm/labsafety.pdf.

2. UVM UVM Campuswide Emergency Response Hazards Report.


http://esf.uvm.edu/labhazards/.

60
3. UVM UVM Laboratory Building Hazards for Cook Physical Sciences, 82
University Pl.
http://esf.uvm.edu/labhazards/chems_in_bldg.php?code=125&showold=0.

4. (a) UCLA Laboratory Hazard Assessment Tool (LHAT).


https://http://www.ehs.ucla.edu/research/lab/lhat; (b) UCLA, Laboratory Hazard
Assessment Tool; (c) UVM UVM Laboratory HCOC Inventory.
https://http://www.uvm.edu/~esf/lab_inv/.

5. (a) UMN Joint Safety Team. http://www.jst.umn.edu; (b) UMN Graduate Students
and Postdoctorates Lead Laboratory Safety Efforts.
http://cse.umn.edu/admin/comm/features/2012_9_19_dowchemical-safety.php; (c)
UMN Joint Safety Team: About. http://www.jst.umn.edu/about.html.

6. Foderaro, L. W., Yale Student Killed as Hair Gets Caught in Lathe. The New York
Times 2011.

7. UVM Guide to Managing Chemical Waste in UVM Laboratories.


http://www.uvm.edu/~riskmgmt/wasteguide.pdf.

8. UVM, Unattended Operations. 2001 Updated 2012.

9. UVM UVM Chemical Hygiene Plan. http://esf.uvm.edu/uvmchp/.

10. Gibson, J. H.; Schroder, I.; Wayne, N. L., A Research University's Rapid Response
to a Fatal Chemistry Accident: Safety Changes and Outcomes. J. Chem. Health Saf.
2014, 21 (4), 18-26.

11. UMN Chemical Safety Audit Checklist.


http://www.dehs.umn.edu/html/checklist.htm.

61
CHAPTER 5: CONCLUSIONS AND NEXT STEPS

5.1. Training Recommendations

Incidents and their repercussions at UCLA, UMN, and UVM were discussed at

length above, but further steps after completing all of the safety adjustments have not yet

been considered. UCLA has made great progress to increase the strength of the safety

culture at the university as described in Chapter 2.1-2 The numerous SOP templates

available for researchers to adapt to their own procedures are immensely useful.2 If the

researchers use them appropriately as a tool to create their own SOPs, the template

library will be invaluable to the research community. Aside from the SOP library, the

increased training and documentation will help to keep everyone safer – those working in

the laboratories as well as any emergency personnel that may need to enter the laboratory

space at any time. The number of trainings received by UCLA affiliates has increased

steadily from 2007 through 2012, as illustrated in Figure 22.

Figure 22: Total trainings received by UCLA affiliates from 2007 through 2012. The blue portion
represents online trainings, and the red portion represents classroom trainings.1

62
Similar to UVM, there is likely some amount of back training students, faculty,

and staff that had not previously received required trainings from 2009 through 2011 as

can be recognized through the large increase in total training participants. In future years,

it may be possible to see a slightly fluctuating plateau of trainings completed, such as

from 2011 to 2012, as it will be only new researchers completing the majority of training

courses required. One aspect of the UCLA training program that is not seen at every

university is a matrix to determine what trainings are required based on what tasks that

specific researcher is performing in the laboratory. The matrix includes different

categories of required trainings for those working with animals, with shop equipment,

responsible for a laboratory, etc.3 At UVM, there are a number of online and classroom

training requirements for every laboratory worker that are clearly indicated; however, it

can be difficult for a researcher to determine which of the more specialized trainings are

also required and for whom these requirements apply.4 A more precise training matrix

would be a beneficial addition to the safety program for researchers to be better able to

complete the necessary requirements.

5.2. Incident Follow Up Procedures and Reporting

Even with all of the progress already made, there are still more steps to be taken

to ensure the safest workplace/laboratory possible. The worst outcome after the prior

incidents would be to become complacent with the safety program. As noted in Chapter

1, there is always room for improvement, features to learn new best practices, and new

63
regulations to follow. One area that could use improvement at UVM is the recording and

tracking of accidents and injuries that happen on campus. The current process of incident

reporting and follow up is outlined in Figure 23.

Figure 23: Reporting process for accidents and injuries at UVM.

The accident or injury goes through the same reporting and follow up procedure whether

the injured person is an employee or a student. After the injured person has received

medical treatment – anything from a bandage to emergency care at the hospital – his or

her laboratory supervisor will complete the report to send in to RMS, ideally within 72

hours of the incident.

After the incident is reported, the Laboratory Safety Coordinator from RMS

assigned to that department does a follow up with the person(s) that had experienced the

injury or accident and likely his or her PI. As discussed in Section 4.3, the follow up

process endeavors to determine the root causes of the accident. Depending on the

incident, other researchers in the laboratory may be included in the discussion to

determine root causes. Follow up may include a simple discussion among researchers

and safety staff, a cooperation to develop a safer procedure, exposure monitoring during

64
experimentation, or a number of other methods depending on the accident or injury in

question, including those described in Section 4.3. The overall goal is to prevent

additional similar injuries or accidents from happening. After the follow up is completed,

the RMS report is sent to another member of RMS for insurance purposes or it is filed. It

is difficult at best to find information about a specific type of injury because there is no

useful way of sorting the reports. A database needs to be created to make the reports—

from the PI as well as the report from RMS staff—useful for tracking purposes and to

determine if any similar accidents happened in the past.

5.3. Final Thoughts

As detailed in Section 3.3 and Section 4.1, UMN and UVM have similar signage

for any operations left unattended. However, UMN has other signage that could be

useful for UVM to mimic. For instance, fillable forms that can communicate more

information to anyone outside of the laboratory could be useful. The availability of one

or multiple forms that can be tailored to a specific reaction, while still following a

particular template would allow for valuable information to be conveyed. Similar to the

SOCs that UMN uses, concise information that can be easily understood by emergency

responders would be beneficial.

The Chemistry Safety Committee was thoroughly discussed in Section 4.1.2,

however, most departments at UVM do not have specific safety committees. If each

department created a committee, events like safety days could be held in other buildings

65
across campus. Though the chemistry department has laboratories with some of the

highest hazards on campus, it is not the only department that would benefit from an

increase in the overall culture of safety. At a minimum, each department should have an

assigned faculty or staff member to oversee the building’s safety needs and coordinate

with RMS to increase the culture of safety.

Ultimately, the safety culture at universities needs to be top priority to minimize

accidents and maximize laboratory effectiveness. If no other reason resonates with

researchers, laboratory accidents impede progress in research. Therefore, universities

need to continue to develop their safety programs and promote safety culture.

5.4. References

1. Gibson, J. H.; Schroder, I.; Wayne, N. L., A Research University's Rapid Response
to a Fatal Chemistry Accident: Safety Changes and Outcomes. J. Chem. Health Saf.
2014, 21 (4), 18-26.

2. UCLA Standard Operating Procedure (SOP) Template Library.


http://www.sop.ehs.ucla.edu.

3. UCLA, Safety Training Matrix. 2015.

4. UVM, Train and Inform Lab Personnel. Updated 2015.

66
CHAPTER 6: COMPREHENSIVE BIBLIOGRAPHY

Banister, M. R. (2005). "An Environmental, Health and Safety Class for Undergraduate
Chemistry Majors." Chemical Health & Safety 12(2): 20-22.

Birkofer, L. and P. Wegner (1970). "Trimethylsilyl Azide." Organic Syntheses 50: 107.

Blum, D. (2013) Bad Chemistry: an Update on the Sheri Sangji Case. Wired

Bracher, P. (2011). UCLA Professor Patrick Harran Charged in Sheri Sangji's Death.
ChemBark.

Bracher, P. (2013). UCLA Chemistry Professor Patrick Harran to Stand Trial. ChemBark.

Bracher, P. (2014). UCLA Professor Patrick Harran Strikes Deal with Prosecutors.
ChemBark.

Buxbaum, E. C. (1934). "Safety in the Chemical Laboratory." Journal of Chemical


Education 11(2): 73-76.

Cal/OSHA (2009). "Citation and Notification of Penalty." from


http://pubs.acs.org/cen/_img/87/i31/Cal-OSHA_UCLA_citation_2009-05-04.pdf.

Chemjobber (2014). ChemBark: Not Enough Punishment for Harran. Chemjobber.

Chen, C.-Y., et al. (2013). "Synthesis, DNA-Binding Abilities and Anticancer Activities
of Triazole-pyrrolo[2,1-c][1,4]benzodiazepines Hybrid Scaffolds." Bioorganic &
Medicinal Chemistry Letters 23(24): 6854-6859.

Christensen, K. (2009). Dealdy UCLA Lab Fire Leaves Haunting Questions. Los Angeles
Times.

Cooper, M. D. and R. A. Phillips (2004). "Exploratory Analysis of the Safety Climate


and Safety Behabior." Journal of Safety Research 35(5): 497-512.

Davison, H. F. (1925). "Safety First in Storing Chemicals." Journal of Chemical


Education 2: 782-783.

DeJoy, D. M., et al. (2010). "Making Work Safer: Testing a Model of Social Exchange
and Safety Management." Journal of Safety Research 41(2): 163-171.

67
DeJoy, D. M., et al. (2004). "Creating Safer Workplaces: Assessing the Determinants and
Role of Safety Climate." Journal of Safety Research 35(1): 81-90.

Dow (2012). "Safe Operation Card (SOC)." from


http://storage.dow.com.edgesuite.net/safety-dow-com/External-SOC-Form_v060612.pdf.

Foderaro, L. W. (2011). Yale Student Killed as Hair Gets Caught in Lathe. The New
York Times. New York.

Gibson, J. H., et al. (2014). "A Research University's Rapid Response to a Fatal
Chemistry Accident: Safety Changes and Outcomes." Journal of Chemical Health and
Safety 21(4): 18-26.

Hill Jr., R. H. (2003). "Getting Safety into the Chemistry Curriculum." Chemical Health
& Safety 10(2): 7-9.

Hill Jr., R. H. (2005). "Strengthening Safety Education of Chemistry Undergraduates."


Chemical Health & Safety 12(6): 19-23.

Hill Jr., R. H. (2007). "The Emergence of Laboratory Safety." Journal of Chemical


Health and Safety 14(3): 14-19.

Keller, E. (1910). "The Hygiene of the Small Chemical Laboratory." Industrial &
Engineering Chemistry 2(6): 246-251.

Kemsley, J. (2010). "Sodium Azide + Acid = Boom." The Safety Zone by C&EN. from
http://cenblog.org/the-safety-zone/2010/04/sodium-azide-acid-boom-2/.

Kemsley, J. (2012). "Dow Chemical Teams Up With Universities On Laboratory Safety."


Chemical & Engineering News 90(44): 21-23.

Kemsley, J. (2014). "C&EN Talks Safety with Minnesota Chemistry Chair William B.
Tolman." Chemical & Engineering News 92(45): 34.

Kemsley, J. (2014). "Explosion Injures University of Minnesota Graduate Student." The


Safety Zone by C&EN. from http://cenblog.org/the-safety-zone/2014/06/explosion-
injures-university-of-minnesota-graduate-student/.

Kemsley, J. (2014). "Lab Death Legal Defense Cost University of California Nearly $4.5
Million." Chemical & Engineering News 92(44): 9.

68
Kemsley, J. (2014). "More Details on the University of Minnesota Explosion and
Response." The Safety Zone by C&EN. from http://cenblog.org/the-safety-
zone/2014/07/more-details-on-the-university-of-minnesota-explosion-and-response/.

Kemsley, J. N. (2009). "Learning from UCLA." Chemical & Engineering News 87(31):
29-31, 33-34.

McGarry, K. A., et al. (2013). "Student Involvement in Improving the Culture of Safety
in Academic Laboratories." Journal of Chemical Education 90(11): 1414-1417.

Morris, J. (2012). "Landmark Worker Death Case Continues Against UCLA Chemistry
Professor." from http://www.publicintegrity.org/2012/07/27/10313/landmark-worker-
death-case-continues-against-ucla-chemistry-professor.

Mulcahy, M. B., et al. (2013). "College and University Sector Response to the U.S.
Chemical Safety Board Texas Tech Incident Report and UCLA Laboratory Fatality."
Journal of Chemical Health and Safety 20(2): 6-13.

Nelson, D. A. (1999). "Incorporating Chemical Health and Safety Topics into Chemistry
Curricula." Chemical Health & Safety 6(5): 43-48.

Sigma-Aldrich (2012). Technical Bulletin AL-134: Handling Air-Sensitive Reagents.

Sigma-Aldrich (2015). "Safety Data Sheet: Sulfuric acid." from


https://http://www.sigmaaldrich.com/MSDS/MSDS/DisplayMSDSPage.do?country=US
&language=en&productNumber=320501&brand=SIAL&PageToGoToURL=https%3A%
2F%2Fwww.sigmaaldrich.com%2Fcatalog%2Fsearch%3Fterm%3Dsulfuric%2Bacid%2
6interface%3DAll%26N%3D0%26mode%3Dmatch
partialmax%26lang%3Den%26region%3DUS%26focus%3Dproduct.

Taton, T. A. and W. E. Partlo (2014). "Chemical Safety: Explosion Hazard in Synthesis


of Azidotrimethylsilane." Chemical & Engineering News 92(43): 2-3.

Trager, R. (2014). "UCLA Chemist Avoids Prison Time for Lethal Lab Accident." from
http://www.rsc.org/chemistryworld/2014/06/chemist-avoids-prison-sheri-sangji-lab-
safety-case-settlement.

Trager, R. (2014). UCLA Spent $4.5 Million on Legal Costs in Sangji Case. Chemistry
World, Royal Society of Chemistry.

Turner, R. G. (1925). "Suggestions to the Teacher of Chemistry on Poisons and Their


Treatment." Journal of Chemical Education 2: 466-471.

69
Turrill, P. L. (1933). "Public Liability and Chemical Education." Journal of Chemical
Education 10(9): 552-555.

UCLA. "Chemical Safety Storage Guidelines." from


https://http://www.ehs.ucla.edu/research/lab/chem/storage.

UCLA "Laboratory Hazard Assessment Tool."

UCLA. "Laboratory Hazard Assessment Tool (LHAT)." from


https://http://www.ehs.ucla.edu/research/lab/lhat.

UCLA. "Standard Operating Procedure (SOP) Template Library." from


http://www.sop.ehs.ucla.edu.

UCLA (2009). "Report to the Chancellor of UCLA Laboratory Safety." from


https://ucla.app.box.com/ehs-report-to-chancellor.

UCLA (2009). "UCLA Laboratory Safety Committee." from


https://http://www.ehs.ucla.edu/research/lab/committee.

UCLA (2012). "Standar Operating Procedure: tert-Butyllithium."

UCLA (2014). "General Services Campus Map." from http://maps.ucla.edu/campus/.

UCLA (2014). "UCLA Chemical Hygiend Plan." from https://ucla.app.box.com/ehs-


chemical-hygiene-plan.

UCLA (2015). "Safety Training Matrix."

UCSB (2015). "Near Miss Reporting." from


http://www.chem.ucsb.edu/about/safety/near-miss.

UDel (2013). "Near Miss Reporting." from


http://www.udel.edu/ehs/generalhs/indhygiene/nearmiss-reporting.html.

UMN (2010). "Chemical Safety Audit Checklist." from


http://www.dehs.umn.edu/html/checklist.htm.

UMN (2011). "U of M to Receive Nearly $17 Million Over the Next Five Years from
Dow Chemical Company." Discover. from http://discover.umn.edu/news/business-law/u-
m-receive-nearly-17-million-over-next-five-years-dow-chemical-company.

70
UMN (2012). "The Safe Chemical Storage Guide."

UMN (Updated 2015). "Dow + U = Lab Safety: A New Safety Campaign for
Laboratories Takes Hold." from http://discover.umn.edu/news/vision-leadership/u-and-
dow-chemical-team-lab-safety.

UMN (Updated 2015). "Graduate Students and Postdoctorates Lead Laboratory Safety
Efforts." from http://cse.umn.edu/admin/comm/features/2012_9_19_dowchemical-
safety.php.

UMN (Updated 2015). "Joint Safety Team." from http://www.jst.umn.edu.

UMN (Updated 2015). "Joint Safety Team: About." from


http://www.jst.umn.edu/about.html.

UMN (Updated 2015). "Safety Initiative Becoming Embedded into Department's


Culture." from http://www.chem.umn.edu/news/news.lasso?serial=642.

USCSB (2010). "Texas Tech University Laboratory Explosion." from


http://www.csb.gov/assets/1/19/csb_study_ttu_.pdf.

USCSB (2011). Experimenting with Danger.

USCSB (2011). ""Experimenting with Danger" Focuses on CSB Case Study on Texas
Tech University Accident; Laboratory Deaths at UCLA and Darmouth." from
http://www.csb.gov/csb-releases-new-video-on-laboratory-safety-at-academic-
institutions/.

UVM. "Segregate and Store Chemicals Safely." from


http://www.uvm.edu/safety/lab/label-and-store - segregate%20&%20Store.

UVM (2001). "Guide to Managing Chemical Waste in UVM Laboratories." from


http://www.uvm.edu/~riskmgmt/wasteguide.pdf.

UVM (2001 Updated 2010). "UVM Chemical Hygiene Plan." from


http://esf.uvm.edu/uvmchp/.

UVM (2001 Updated 2012). "Unattended Operations."

UVM (2012). "Laboratory Health and Safety." from


http://www.uvm.edu/policies/riskmgm/labsafety.pdf.

71
UVM (2015). "UVM Campuswide Emergency Response Hazards Report." from
http://esf.uvm.edu/labhazards/.

UVM (2015). "UVM Laboratory Building Hazards for Cook Physical Sciences, 82
University Pl." from
http://esf.uvm.edu/labhazards/chems_in_bldg.php?code=125&showold=0.

UVM (2015). "UVM Laboratory HCOC Inventory." from


https://http://www.uvm.edu/~esf/lab_inv/.

UVM (Updated 2015). "Train and Inform Lab Personnel."

Walker, F. (1934). "Safety in Chemical Laboratories." Journal of Chemical Education 11:


506-509.

Wiss, J., et al. (2007). "Explosion and Decomposition Characteristics of Hydrazoic Acid
in the Gas Phase." Organic Process Research & Development 11(6): 1096-1103.

72