by Ed Acheson
Department of Chemistry
Millikin University, Decatur, IL
This case was inspired by the story of Willie Jones (1). Mr. Jones was arrested after drug-sniffing dogs detected cocaine in a suitcase he was carrying. The suitcase contained $9000, which was seized by Drug Enforcement Agency agents under the federal Racketeer Influenced and Corrupt Organizations (RICO) law. RICO allows the government to seize assets, such as money, that are suspected of having been used to commit crimes. Mr. Jones later sued the government for the return of his money. Mr. Jones contended that cocaine contamination of paper currency is so widespread that the mere presence of cocaine is not a sign of criminal activity. Federal Judge Thomas A. Wiseman, Jr. agreed with Mr. Jones and ordered his money returned (1).
I have used this case successfully in a non-majors science course. In this course, the case was used as a fun way to introduce students to lab and lab instrumentation. The student lab results were used as the basis for our class discussions, although information on cocaine contamination given in references (2-4) could substitute for student-generated data.
I. Experimental Design
Scientists generally plan their experiments, even if the results turn out not to be what was expected! Scientists carefully consider what they hope to determine and the variables involved in the process. They then lay out the procedures they will follow in the lab, perform the experiments, and collect and analyze the data. The results from one set of experiments often suggest other experiments to perform, so the process repeats itself in a cyclical manner.
The experimental procedure in this case study is relatively straight-forward, if you have a GC/MS available. There are, however, several variables your students (and you) may want to explore. (Even if you do not have a GC/MS, your students could still design the experiments they would perform if they could. This is still a valuable introduction to experimental design.) In addition to the example questions given in the case handout, your students might consider such variables as:
Once students have collected their data, or collected data from literature sources, you can begin discussing their interpretation of the data. This discussion can take a variety of forms, depending upon the level of the students and the data they have collected. One topic you may want to emphasize in the discussion is "certainty": how "certain" or "confident" are you in your results? If 80% of the bills the students tested are contaminated, does that mean 80% of all bills in circulation are contaminated? Depending upon your personal inclination, this could lead to a discussion of statistics and/or sampling.
II. RICO
Congress passed the Racketeer Influenced and Corrupt Organizations (RICO) law in 1961. The law can be found in United States Code TITLE 18 - CRIMES AND CRIMINAL PROCEDURE PART I - CRIMES CHAPTER 96 - RACKETEER INFLUENCED AND CORRUPT ORGANIZATIONS (RICO) Section 1961. Definitions. Section 1962. Prohibited activities. Section 1963. Criminal penalties. Section 1964. Civil remedies. Section 1965. The text of the complete RICO law can be found at the following commercial web site (accessed November 1999): http://dresch-jheon.simplenet.com/empower/18uscc96.htm
Here is a small section from the law that states Congress’ purpose in passing the law:
CONGRESSIONAL STATEMENT OF FINDINGS AND PURPOSE
Section 1 of Pub. L. 91-452 provided in part that: "The Congress finds that (1) organized crime in the United States is a highly sophisticated, diversified, and widespread activity that annually drains billions of dollars from America's economy by unlawful conduct and the illegal use of force, fraud, and corruption; (2) organized crime derives a major portion of its power through money obtained from such illegal endeavors as syndicated gambling, loan sharking, the theft and fencing of property, the importation and distribution of narcotics and other dangerous drugs, and other forms of social exploitation; (3) this money and power are increasingly used to infiltrate and corrupt legitimate business and labor unions and to subvert and corrupt our democratic processes; (4) organized crime activities in the United States weaken the stability of the Nation's economic system, harm innocent investors and competing organizations, interfere with free competition, seriously burden interstate and foreign commerce, threaten the domestic security, and undermine the general welfare of the Nation and its citizens; and (5) organized crime continues to grow because of defects in the evidence-gathering process of the law inhibiting the development of the legally admissible evidence necessary to bring criminal and other sanctions or remedies to bear on the unlawful activities of those engaged in organized crime and because the sanctions and remedies available to the Government are unnecessarily limited in scope and impact."
"It is the purpose of this Act (see Short Title note above) to seek the eradication of organized crime in the United States by strengthening the legal tools in the evidence-gathering process, by establishing new penal prohibitions, and by providing enhanced sanctions and new remedies to deal with the unlawful activities of those engaged in organized crime."
Here is a synopsis of the law taken from Black's Law Dictionary, 6th Edition (West Publishing Company, Minnesota, 1990):
"RICO laws - Racketeer Influenced and Corrupt Organizations laws. Federal and state laws designed to investigate, control, and prosecute organized crime. 18 U.S.C.A. Section 1961 et seq. Both criminal prosecution and civil actions may be brought under RICO statutes. Federal RICO laws prohibit a person from engaging in activities which affect interstate or foreign commerce, including: (1) using income received from a pattern of racketeering to acquire an interest in an enterprise; (2) acquiring or maintaining an interest in an enterprise through a pattern of racketeering; (3) conducting or participating in the affairs of an enterprise through a pattern of racketeering; and, (4) conspiring to commit any of the above offenses. To establish a prima facie RICO claim, a civil plaintiff or prosecutor must allege the existence of seven elements: "(1) that the defendant (2) through the commission of two or more acts (3) constituting a 'pattern' (4) of 'racketeering activity' (5) directly or indirectly invests in, or maintains an interest in, or participates in (6) an 'enterprise' (7) the activities of which affect interstate or foreign commerce." 18 U.S.C.A. Section 1962. Moss v. Morgan Stanley, Inc., C.A. N.Y., 719 F.2d 5, 17."
Criminal penalties under RICO include a fine or imprisonment, or both, along with forfeiture of any property acquired during the course of the illegal racketeering activity.
Many states have passed RICO laws, and both criminal and civil penalties can be imposed depending upon the particular RICO law. RICO laws have been applied to a number of cases outside organized crime racketeering. When your students do a web search for "Racketeer Influenced and Corrupt Organization law," they will come up with a wide variety of lawsuits involving RICO. (For example, the Detroit Daily News sued striking unions under civil RICO laws, and pro-choice groups have sued pro-life groups under RICO for conspiring to forcibly close abortion clinics across the country.) The cases your students find may make interesting discussion questions and cases outside of this one!
III. Sampling
When a chemist takes a sample to be analyzed, he or she has to be sure that the sample is both: (1) Representative of the bulk material being sampled; and (2) Homogeneous. A homogeneous sample is simply one that is uniform throughout. In this experiment, we have to assume that the bills the students sample are in fact homogeneous; i.e., all paper currency is made of the same material, and all bills are identical when they are printed. Of course, the bills the students collect will vary in their dirt and grime content depending upon their age and how much they have been handled.
A bigger problem in this case study is obtaining a representative sample. In 1997 over 4,600,000,000 one-dollar bills were printed. With that many bills having been printed, how can we be sure that the bills the students sample are representative of all of the bills printed? We can't. Instead, we have to assume that the distribution of cocaine contaminated bills throughout the United States follows a binomial distribution. We also have to assume that the bills the students sample have in fact been collected randomly, and since they have been collected randomly they are representative of all the bills in circulation. Statisticians might not agree with our assumptions, but for the purposes of this experiment, these are reasonable assumptions.
There are two interesting questions students might raise with regards to their experimental results:
Assuming the distribution of contaminated bills is random (which we are), these questions can be addressed using the idea of the "confidence interval."
In its simplest form, the confidence interval describes the range around the experimental mean where we would expect to find the "true" value to fall with a given probability, or certainty. The confidence interval can be calculated from the following formula:
Confidence interval =
where 
= your experimental result
= 1 - 
n = number of data points in the set
= a statistical factor called the Z-score.
For a sample distribution that follows a binomial distribution, the value of z at the 95% confidence level is 1.96. Knowing the experimental value of p (the percentage of contaminated bills) and the number of bills sampled, we can calculate the confidence interval for our data.
For example, we recently tested 10 bills in our lab; 80% of the bills tested positive for cocaine. The confidence interval for these data is:
Confidence interval = 
= 0.80 
0.25
Thus, we can say that we are 95% confident that the number of bills contaminated with cocaine lies somewhere in the range 55-105% (0.80 0.25 x 100). In actuality, of course, we cannot have more than 100% contamination. Our confidence interval, therefore, tells us we have 55-100% contamination. This is a relatively broad range. Your students might have differing interpretations of the significance of this number. Even though it is a broad range, your students might conclude that a significant portion of the bills in circulation are contaminated. Others might look at the range and conclude that it is too wide to be of significance. This might be an interesting lead-in to a discussion of interpretation of results. How can different scientists bring opposite interpretations to the same experimental results? This could be an example of this dilemma for your students. You might want to emphasize this point by having your students put on a "mock trial" of this case (see below).
Suppose your students tested 100 bills, and found 80% of them contaminated. In this case, the confidence interval is:
= 0.80 0.08
Now, in this case, we can say that we are 95% confident that the number of bills contaminated with cocaine is somewhere between 72% - 88%.
This is a much smaller range, and might lead some of your students to conclude that more bills must be tested to give more reliable results. We can use the confidence interval to help us answer the second question posed above: How many bills would I have to test to give me a certain, specified error in my results? Unfortunately, there is no definitive answer to that question, since the number of bills to be tested depends mathematically on both the error we want in our measurement, and the number of bills that test positive. Instead, let's illustrate the general principle with some examples. The general equation in this case is:
Measurement error = 
where p, q, n, and z have the same definitions as before.
1. Most public opinion polls tell you the uncertainty in the result is "2%" or "3%". Let's use 2% as our "uncertainty"; that is, the error in our results is 0.02. A "conservative" approach is to also say that 50% of the bills tested test positive for cocaine; in other words, p and q in the equation above are both 0.50. If we substitute these values into our measurement error equation, we get
measurement error = 0.02 = 
where n = the number of bills we would have to sample to have a 2% error in our results (again, assuming 50% of all bills are contaminated). Solving for n gives a value of 2401. In other words, if we want to be 95% confident that the percentage of bills in circulation that are contaminated with cocaine falls in the range 48%-52%, we would have to test 2401 bills.
2. Suppose we are willing to accept a 10% error in our measurement. If the number of bills that tested positive is again assumed to be 50%, the number of bills that we would have to sample is given by
measurement error = 0.10 =
n in this case = 96. Again, in other words, if we want to be 95% confident that the percentage of bills in circulation that are contaminated with cocaine falls in the range 40%-60%, we would have to test 96 bills.
3. How about the situation in our lab? 80% of the bills tested in our lab were positive for cocaine. If we wanted to be 95% confident that 80% 2% of all the bills in circulation were contaminated with cocaine, how many bills would we have to test?
measurement error = 0.02 = 
Solving for n gives 1536.
Depending upon the percentage of bills that test positive experimentally, and the margin of error you can tolerate in your results, students might have to sample as few as 96 bills, or as many as 2401. If we collected one bill from every student and employee at Millikin University, we could collect 2401 bills. But how long and at what laboratory cost would it take to test this number of bills? You might have students work out the cost of running this analysis. Business and accounting majors in particular should find this an interesting exercise.
Another interesting point to note is that the number of bills that we have to sample to achieve a certain measurement error is independent of the total number of bills in circulation. It does not matter if 4 billion, or 4 million, or 400,000 bills are in circulation, we would still have to sample "just" 2401 in our lab to be 2% certain of our result. Students might find it interesting to compare this number to the number of families sampled by A. C. Nielsen to determine television ratings. There are approximately 98 million TV households in the United States. Nielsen samples around 5000 of those households to determine the TV ratings (5).
V. Cocaine (6-8)
For thousands of years, people have known that eating or chewing the leaves, bark, or roots of certain plants causes physiological changes in the body. These changes range from curing disease to a feeling of euphoria to addiction to death by poisoning. Beginning in the 19th century, chemists started to isolate the plant compounds responsible for these effects. The plant compounds were known as alkaloids (alkaline-like) because of their basic behavior toward acids. Besides cocaine, other common alkaloids include nicotine, caffeine, atropine, morphine, quinine, strychnine, and scopolamine.
The alkaloids are all basic because chemically they are all amines. Amines are compounds that are derivatives of the base ammonia, NH3. One, two, or all three of the hydrogen atoms on ammonia can be replaced by organic constituents to form amines. The nitrogen in ammonia has an unshared pair of electrons that give the compound its basic nature.
Cocaine is found in the leaves of the South American coca bush. Cocaine reduces fatigue, produces euphoria, and gives a feeling of immense power. It produces a powerful and fast addiction. Sigmund Freud is considered to be the first person to use cocaine medically. He investigated the use of cocaine to treat his patients. He convinced a fellow physician, Karl Koller, to use cocaine as an aid to break his addiction to morphine. Koller succeeded, but at the price of becoming addicted to cocaine. Koller recognized the anesthetic value of cocaine, and used it medically as a local anesthetic. Cocaine was shortly thereafter used as a dental anesthetic, but because of its toxicity, was replaced first by Novocain and now by Xylocaine.
Pure cocaine is neutral; it is typically isolated from the leaves as cocaine hydrochloride salt. The structure of cocaine hydrochloride is
Cocaine hydrochloride is a salt with properties similar to those of sodium chloride. It is readily soluble in water, is stable towards heat, and does not vaporize readily. Reacting the salt with base converts the salt to neutral base, often called "free base." During this chemical reaction, the neutral base precipitates as a white solid sheet that cracks into lumps or "rocks." The formation of "rocks" as the solid cracks gives rise to the name "crack cocaine" or simply "crack." Unlike the salt, pure cocaine readily vaporizes, and inhaling the vapor produces a more intense "high" than that of the salt.
Cocaine works in the brain by blocking the reuptake of dopamine from the nerve synapse. Numerous studies have shown that dopamine is the neurotransmitter responsible for feelings of reward and satisfaction. The higher the concentration of dopamine in the brain, the greater the reward - the "high" of drug addiction.
VI. Cocaine Contamination
How does money become contaminated with cocaine in the first place? The process begins, naturally enough, with drug dealers and users. Dealers may have cocaine on their hands, which they spread to money. Users, too, may contaminate money from cocaine on their skin. Contamination also occurs when people use rolled-up bills to snort cocaine. According to Argonne National Laboratory chemist Jack Dermirgian (2), "the paper used for US money actually has a saw-toothed fiber with spaces between the fibers." When cocaine falls on currency, folding and wrinkling the bill breaks the cocaine crystal into smaller particles, which become trapped in the spaces between the fibers. But users and dealers alone cannot account for the widespread contamination of money - there must be another mechanism acting to spread the contamination. The prevailing theory is that money sorters at banks and ATM machines spread the contamination (9). Cocaine is picked up by the belts and rollers in these machines from contaminated bills. When clean bills pass through the machines, they in turn are contaminated.
VII. GC/MS
Gas chromatography (GC) is a powerful technique for separating volatile and semivolatile compounds in complex mixtures. It cannot, however, identify the isolated components. Mass spectrometry (MS), on the other hand, provides detailed information on the structure of individual compounds, allowing chemists to exactly identify them. It cannot, however, isolate the components of a mixture. The combination of the two techniques gives the chemist a powerful tool for identifying and separating the constituents of complex mixtures.
GC/MS is the basis for many official EPA methods. It is widely used for the quantification of pollutants in drinking water and wastewater. It is used for the identification and quantification of drugs and drug metabolites in blood and urine. It is also widely used in forensic science and pharmacology.
GC/MS is limited to compounds that have vapor pressures above 10-10 Torr. It is difficult to distinguish between some isomers and the position of substituents on ring compounds using GC/MS. Although the price of instrumentation for GC/MS has dropped considerably in the past 10 years, it is still a relatively expensive technique, with instrument prices starting at around $50,000.
An excellent source for the applications, advantages and limitations of GC/MS is given in ref. (10). Other useful information on GC/MS can be found in textbooks for a course in Instrumental Analysis, such as ref. (11).
CLASSROOM MANAGEMENT
I have used this case study in our non-majors science course. The non-majors course has a different format than our typical class. Class sizes are generally 25-50. The class periods are two hours each, meeting twice a week. Each two-hour period incorporates lecture and/or lab, as the instructor sees fit. When I have lab activities planned, I usually have half the class attend a given class period if the lab will take more than an hour; if the lab activity will take an hour, I have half the class attend the first hour, and the other half attend the second hour.The specific class I used this case study in had 35 students. I introduced the case study by asking students if they had ever been stopped for a traffic violation (most had been). We briefly discussed how it felt to be stopped, what happened (warning, ticket, court appearance?), etc. Then I asked students to imagine how it must feel to be arrested on a more serious charge. After a brief discussion, I broke the class into groups of 5-6. I gave each group "Part 1" of the case as a handout and asked them to read through it. I then had each group decide how they would come up with the answers to the discussion questions. Each group had to turn in a written summary of how they were going to answer the questions or which individual in the group would be responsible for which question(s). I also asked each member of the group to write their a priori answer to question 10. I asked the groups to do this in order to make sure they were using their discussion time (somewhat) productively, and to get a personal commitment to the outcome of the case in hopes that this would provide some incentive to research the questions. We then reconvened as a whole and discussed sources they could use in their research. This was completed in one period.
For the second class period, students reconvened in their groups and discussed their answers to the questions. While they were doing this, I met individually with each group to discuss the experiment(s) they designed. I made comments and suggested changes as appropriate, but mainly I wanted to make sure the students had reasonable experiments to perform given our equipment (I suggested few changes--mainly cutting back on the number of bills they wanted to test). After I met with each group, half the students went to the lab and performed their experiments while the other half waited in the classroom. (The extraction is simple and straight-forward and takes only a few minutes.) It took about 30 minutes for the first group of students to finish the lab work, and then the second group went into the lab. After I collected all the student samples, I ran them on the GC/MS and posted the results outside my office. I asked students to look at the results before the next class period (some did, some didn't).
I began the third class period by having students return to their groups and discuss the answers to their questions as well as the results of their experiments. After this we reconvened as a whole and continued the discussion.
We spent a total of 4-5 hours on this case study. If you have a more traditional course format (three 1-hour lectures and one lab period), you could introduce the case during the first period and meet with the groups during the second period. After all the students had performed the lab work and the results had been compiled, you could discuss the case and results in a third period.
Mock Trial Variation
Your students might enjoy debating this case in the form of a mock trial. Divide the class into 3 groups: two groups of 5, with the rest of the class comprising the third group. One group of 5 can serve as the "prosecution team" and the other group of 5 can serve as the "defense team." Each group should develop a case for their side of the trial. Each group should assign a lead attorney and 1-2 members of the group to be "expert witnesses." Each team should develop a list of questions to ask their own expert, as well as a list of questions to ask the opposing experts. The teacher can be the judge for the case or assign another student to act as judge. The third group acts as the jury, and is responsible for researching proper courtroom procedures for the trial. You can allow as much time as necessary to complete the "trial," but one 50-minute class period should be sufficient time for each side to make its case and for the jury to reach a decision.
We have performed the lab portion of this case study with student groups of up to 25. The lab work consists of two parts: 1. Extraction of the cocaine from the currency; and 2. Analysis of the extract using a GC/MS.
We have found that Kimble glass vials #60958A-6 and vial caps #73802-24400 (both available from Fisher Scientific) are ideal for this experiment.
Part 1. Extraction.
Depending upon how much time you have available, you can run a "quick and dirty" extraction or a more complete, but time consuming, extraction. The quick and dirty procedure is given in the case study handout and is repeated here.
The alternative procedure given below is a more thorough extraction process and gives better results. It does, however, take more time and equipment and is best suited for groups of 10 or less. If you have enough hot plates available and enough hoods, you can use this procedure with larger groups of students. If you do not have a convenient source of nitrogen available, the nitrogen drying step can be safely omitted with no loss of accuracy. Blowing nitrogen across the methanol extract will help speed the evaporation process.
More complete extraction.
Part 2. Analysis
1000 ppm cocaine standard in methanol is available from Sigma-Aldrich (product number C 1528). The standard is supplied in 1-mL vials, and is available without a DEA license.
We use a Shimadzu QP-5050A GC/MS in EI mode equipped with an AOC-20 autosampler to analyze the methanol extracts for cocaine. The chromatographic conditions are as follows:
Column temperature program:
Under these conditions, cocaine has a retention time of 7.3 minutes. The mass spectrum of cocaine has a parent peak at m/z = 303 and a base peak at m/z = 182. The spectrum also has significant peaks at m/z = 82, 105, and 122.
When we perform a "library search" on student samples, the search software will identify the peak at retention time 7.3 minutes as cocaine, as long as the cocaine peak is reasonably large. This is positive identification for cocaine. At lower concentrations of cocaine, the software mis-identifies the cocaine peak. If the student extract has a peak at retention time 7.3 minutes, and the mass spectrum of the peak matches that of cocaine (by inspection), we say the sample tested positive for cocaine. A typical chromatogram and mass spectrum for a 100 ppm cocaine standard is shown below.
APPENDIX 1
Literature Data on ContaminationI. From ref. (3)
All bills tested were $1.00
|
City |
Number of positives |
|
Baltimore |
9 |
|
Miami |
3 |
|
Chicago |
7 |
|
Honolulu |
10 |
|
Kansas City |
9 |
|
Las Vegas |
9 |
|
Los Angeles |
9 |
|
Minneapolis |
8 |
|
Spanish Fort |
9 |
|
Ft. Wayne |
9 |
|
Pittsburgh |
4 |
|
Yellowstone |
5 |
|
Whitefish |
7 |
|
Portsmouth |
10 |
N = 10 for all cities except Miami, where N = 6.
N = 136 total; 108 positives for 79.4% positive
II. From ref. (4)
|
City |
Number of positives |
|
Rockford (Ten $20 bills tested) |
10 |
|
Chicago (Four $1 bills tested) |
3 |
Acknowledgements:
