CHEM 321: EXTRACTION AND PURIFICATION OF CAFFEINE Structurally complex organic chemicals sometimes can be isolated easily from natural sources. More often, a complex compound of interest (an insect pheromone, for example) cannot be found in large quantities, but isolation is still of interest in order to characterize its structure and perhaps devise a method of synthesis. The purpose of this experiment is to learn some practical isolation, purification, and characterization techniques used in many organic chemistry laboratories. In this experiment, the focus techniques are: liquid-solid extraction from a biological source, vacuum filtration to separate any precipitates from liquid, liquid-liquid extraction to separate the compound of interest from contaminants, distillation to remove solvent, and sublimation to purify the product. In addition, the technique of thin-layer chromatography is demonstrated as a way to characterize (i.e., test the identity and purity of) the recovered compound. To facilitate learning these useful organic chemistry laboratory techniques, caffeine, an alkaloid which acts as a stimulant, will be isolated from tea leaves. Its structure is shown below. Caffeine also occurs in kola nuts, and thus is found in cola soft drinks. Two related compounds, theophylline and theobromine, are also found in tea leaves and likely co-purify with caffeine. (They are also found in cacao beans, the source of chocolate.) O H3C N O CH3 N N N CH3 caffeine (1,3,7-trimethylpurine-2,6-dione) O H3C O N H N N N CH3 theophylline (1,3-dimethyl-7H-purine-2,6-dione) O HN O CH3 N N N CH3 theobromine (3,7-dimethylpurine-2,6-dione) Solvent Extraction The first isolation technique to be used is solvent extraction. A desired compound may be dissolved or suspended in an aqueous solvent. In order to recover it, an organic solvent that is not miscible with water is added to the aqueous mixture in a separatory funnel. The desired compound is usually more soluble in the organic solvent than in water. As the two solvents are shaken together, the desired compound becomes distributed between them according to its preference (described by the distribution coefficient or partition coefficient). In general, very non-polar compounds will favor the organic solvent, and very polar compounds and salts will favor the aqueous phase. Vigorous shaking produces small droplets, which significantly increases the area of contact between the two phases and speeds equilibration of the desired compound between them. The shaking should be as vigorous as possible without causing formation of an emulsion. An emulsion is a mixture of immiscible solvents which refuses to separate into two distinct phases. If an emulsion forms, one can try several things to cause it to separate. Sometimes gentle swirling or mixing with a stirring rod will convince the bubbles of solvents to join their own kind. Saturation of the aqueous phase with a salt sometimes changes its density enough, or requires enough water for solvation, to make it let go of the organic solvent. If all else fails, centrifugation will usually work, but this means decanting into centrifuge tubes and then careful transfer back to the separatory funnel. The proper way to use the separatory funnel is explained here and will be demonstrated in the lab. Before adding anything to the funnel, be sure the valve at the bottom can be turned and is closed. To the funnel add (a) the solution containing the desired compound and (b) about 10-20 mL of a different solvent which is immiscible with the original solvent, and in which the desired compound is very soluble. The new solvent (b) is called the extracting solvent, and the process of shaking is called "extracting". One shakes several times with small portions of fresh extracting solvent since this is more effective than one extraction with a large amount. Stopper the funnel and press the head of the stopper against the palm of one hand. Invert the funnel, shake once, and then immediately vent it (still upside down with stopper pressed into hand) by opening the valve. Be sure the outlet points into your hood! It is important to vent the funnel because many organic solvents have high vapor pressures. Thus, as the organic solvent vaporizes the pressure builds up inside the funnel. This can cause the stopper and perhaps hazardous chemicals to blow out. Repeat the process of shaking and venting until there is no more pressure build-up. Then shake for about 15 seconds and return the funnel to its upright position, resting it in a suitable iron ring. Immediately put a narrow strip of paper between the stopper and its socket in the funnel to prevent inside pressure from popping it out sometime later. This also helps when it's time to drain the lower layer! Always be sure you know which layer is water and which is organic. Most often the lower layer is water, but in this experiment the chloroform solvent is densest and so the organic layer is on the bottom. If you are ever in doubt, it's easy to identify the organic and water layers. Just draw off a little of the lower layer into a test tube containing a small amount of water and mix. If you do not see two layers, the bottom layer is the aqueous, or water, phase. If you do see two layers, the bottom is the organic phase. The contents of the test tube can be poured back into the funnel. EXPERIMENTAL SECTION Isolation of caffeine CAUTION: Chloroform is a suspected carcinogen; avoid breathing vapors (use hood) and skin contact. Add about 350 mL of tap water to a 600 mL beaker and begin heating it. Get a long magnetic stir bar and write your name on the blackboard (erase it when you return the stir bar). Drop the stir bar into the beaker, start stirring, and cover with a watch glass. Weigh an empty 250 mL beaker, then empty the contents of ten tea bags into it and reweigh. Record the net weight of tea. To the hot water in the 600 mL beaker add 12 g of calcium carbonate or 9 g of calcium hydroxide powder, about a tablespoonful of rock salt, and the tea leaves. Cover the beaker with a watchglass, bring to a boil and simmer gently while stirring fast (with stir bar) for about 15 minutes. This brewing procedure swells the tea leaves and extracts caffeine and several other substances into the water. The purpose of the CaCO3 or Ca(OH)2 is to convert any extracted tannins or other acids into anions (negatively charged ions). In leaves these acids exist as electrically neutral polar molecules, and tend to be soluble in both water and CHCl3, complicating the purification and increasing the tendency to form an emulsion in the next step. As anions they are not soluble in CHCl3, and this helps avoid an emulsion. The salt seems to cause precipitation of some substances that tend to clog the filter paper in the next step and also helps avoid formation of emulsions. While the tea is boiling, prepare a 500 mL vacuum flask with a large Büchner funnel. Put three properly fitted sheets of filter paper in the funnel, wet them, and pull a vacuum to seal them to the funnel. Get about 50 mL of Celite (powdery diatomaceous earth, a filter aid) and add enough tap water to make a thin slurry. With the vacuum on, swirl the suspension strongly and pour it all at once and rapidly into the Büchner funnel to create a more or less uniform pad of Celite, but release the vacuum as soon as the pad appears dry. The purpose of this pad is to prevent the filter paper from clogging when you filter the tea. Discard the Celite wastewater that came through the filter. Cool the tea in a sink of cold water until the beaker feels about the same temperature as the cold water. If the tea is too warm, it will cause a lot of foaming in the next step. Let the solids settle and avoid stirring them up. Turn on the vacuum and pour the supernatant liquid through the Büchner funnel. Watch out for foaming in the vacuum flask. If any occurs, bleed a tiny bit of air into the side arm. With a little practice you can easily control the level of foam. If the apparatus clogs, decant the contents of the funnel back into your beaker, clean the funnel, make a new pad, and try again (you don't need to refilter what came through already). Put a 500 mL separatory funnel into an iron ring on a ring stand. Pour in the tea solution and add about 25 mL of CHCl3. Extract caffeine into the organic layer using the technique demonstrated. Shake as vigorously as you can while still getting decent separation of layers — feel free to ask for advice. Drain the lower layer into a 125 mL Erlenmeyer flask. Add 25 mL of fresh CHCl3 to the separatory funnel and shake again. Drain the lower layer into the same 125 mL flask and discard the upper layer. Wash the separatory funnel with tap water. Pour in the saved CHCl3, add about 20 mL of tap water, and shake (this is to remove any pigments or other impurities that prefer water to CHCl3). Drain the lower layer into a dry 125 mL flask and check carefully for visible water droplets. Ask an instructor to verify that it is in fact dry. Once dry, add enough anhydrous calcium chloride to thinly cover the bottom of the flask. This will absorb the dissolved water. Cork the flask (cork, not rubber), wrap the neck and cork completely with Parafilm (as shown by instructor or TA), label using tape with name, date, section, and what it is, and store in your second lab drawer until next period. [This is the end of Week 3. There is no assignment due next week, but you could start writing up your skeleton report while this week’s half is still fresh in your mind.] Next period, set up for simple distillation (see the example in lab) and drive off most of the solvent. Stop heating when a puddle about the size of a quarter remains in the boiling flask. If you have only dry crusts left, you can dissolve them in some of the chloroform you just distilled. [In general one should not distill to dryness because some solvents (such as diethyl ether) can contain peroxides that detonate when dry and hot. In this experiment that is not a problem.] Swirl the boiling flask to dissolve the solid residue (add a little distilled CHCl3 if needed). Use a Pasteur pipet to transfer about two drops of this solution of crude caffeine to the first of 3 clean tiny (smallest available) test tubes for later analysis by thin-layer chromatography (TLC), and the rest to a clean 125 mL vacuum flask. Seal this flask with a stopper containing a test tube (the "cold finger" to be described below). Clamp it to a ring stand in such a way that the side arm and clamp arm point in about the same direction. Pull a vacuum (aspirator ALL THE WAY ON) to evaporate the solvent (the residue must become completely dry). Caffeine can be purified by sublimation in vacuo since it has an appropriate vapor pressure at a temperature below its melting range of 234-237° C. Take the ring stand with vacuum flask to a place where you can both pull a vacuum on it and have access to a gas burner. Refer to the picture on the next page. Fill the test tube in the stopper with ice chips (wipe up any liquid water on the stopper) and insert it into the flask neck. This will act as a cold finger to condense the caffeine vapors in the next step. Hold a heat shield so as to protect the rubber vacuum hose from heat beneath it. Apply vacuum to the flask. Heat the flask with a flame, starting on the sides and working toward the bottom. The objective is to keep all sides and bottom hot - otherwise the caffeine will condense on the flask walls instead of on the test tube. As soon as all the light colored material is on the test tube, cease heating. Very carefully release the vacuum (so your caffeine doesn’t get knocked off the cold finger by a rush of air), take the ring stand back to your desk, withdraw the test tube and immediately empty the ice and water. Stand the test tube on its open end while your tare a piece of weighing paper (use an analytical balance in the instrument room). Scrape the purified caffeine onto the paper and weigh again (record result). Put 1 - 2 crystals (a barely visible amount) in a 2nd tiny test tube to use later for thin layer chromatography (TLC). Put the same amount of authentic caffeine in a third tiny test tube. Ice chips in test tube (cold finger, outside of tube must be dry) Rubber hose to vacuum Heat shield Continuously flame entire surface Analysis of crude and sublimed caffeine Thin layer chromatography (TLC) is one of several analytical methods that can be used to test the purity of a given sample. Therefore, it is one method to test the effectiveness of purification. All types of chromatography depend on distribution of solutes between a stationary solid phase and a moving fluid phase. There are two factors that affect the results in TLC: 1) solvent type and 2) developing plate (or strip) type. The ideal solvent (or mixed solvent) is one that will separate the desired compound from contaminants based on polarity. Suitable solvent systems are chosen based on trial and error. In this experiment, ethyl acetate and ethanol are used in a 9:1 ratio. A variety of developing plates (or strips) are available for TLC. In this experiment, powdered silica (SiO2) on a plastic backing is used as the solid support medium. Since silica is very polar and the plate (or strip) is exposed to air, it attracts a tightly bound layer of water, and polar molecules are strongly attracted to water. A less polar solvent migrates up the solid support by capillary action. Each individual compound in the sample that is put onto the strip (in this experiment, caffeine and possibly other compounds) initially adsorbs to the solid phase at the starting position - the beginning of a lane of travel. Then, as the solvent passes by, each different solute has a choice of whether to interact with the very polar stationary phase or the less polar moving phase. Each compound will partition between the two in accord with its unique properties. In general, polar molecules will prefer the polar stationary phase, while less polar molecules will prefer the less polar moving phase. Thus, separation occurs based on polarity, with molecules of low polarity moving faster in the lane than molecules of high polarity. If the solvent is very polar, then the solutes will not experience much difference in choice between the stationary and solvent phases. Thus, there will be little reason for any solute to move much and the separation will be poor. Less polar solvent systems create better separation between relatively non polar molecules and polar molecules, since polar molecules tend to be more attracted to the stationary phase and non polar molecules tend to be more attracted to the relatively non polar solvent. CAUTION: Ethyl acetate/ethanol solvent is flammable; do not use near open flame. Whenever you use an open flame, always check that no flammable organic solvents are nearby. Thin-layer chromatography: Test the purities of the crude and sublimed caffeine, along with a control (authentic caffeine), by subjecting them to thin layer chromatography. Use silica-coated plastic strips. The coating is extremely fragile, so handle the strip only by the edges (like a photograph). Carefully scribe a very light pencil line across one end of the strip 1 cm from the bottom. Make small crossmarks where you plan to apply the 3 samples. They should be spaced equally from each other AND from the sides of the strip. Dissolve each of the three caffeine samples (approx. 2 crystals for each) in about 2 drops of the distilled CHCl3. Get a capillary tube and dip one end into the test tube containing the authentic caffeine solution. The CHCl3 solution will partly fill it. Working quickly, spot the strip by lowering the capillary vertically until it just touches the silica surface at one of your marks (don’t rip up the coating). The liquid will be rapidly drawn out. Next, spot the sublimed material, and finally the crude. Make the spots as small as possible. A single application of each solution is enough. Overloading is a common problem. Grasp the top of the strip (the end farthest from the spots) with forceps and lower it into a staining jar previously equilibrated with the developing solvent (composed of 9 parts of ethyl acetate plus 1 part of ethanol). The edges of the strip must NOT touch anything or the solvent will not travel up uniformly! Replace the lid on the jar, and do not open it unless necessary. IMPORTANT: the reservoir of developing solvent must be below the sample spots, otherwise they will just dissolve in the solvent and your strip will be blank. Develop the strip until the solvent front approaches the top of the strip. The solvent must not reach the top of the strip, otherwise the Rf calculation will be invalid. Remove the strip and immediately mark the position of the solvent front (after the strip dries it is sometimes difficult to determine where the solvent front was located). Don't forget to replace the cover on the staining jar - the atmosphere inside must stay saturated with solvent. Allow the strip to air dry (many chemists use a blow drier so they don’t have to wait around). Observing the result: There are at least two different ways that allow an observer to visualize the results of TLC: 1) staining with a dye, and 2) using UV light. In this experiment, short wavelength UV light will be used. DANGER - UV radiation is damaging to the eye. Use short wavelength UV light to locate the position of the caffeine (the major spot in each lane) and any other substances (other spots in the same lane). Outline all spots with a sharp pencil. [If there is a long streak, too much material was loaded onto that lane. Another strip may be necessary using a more dilute solution. Ask your instructor or TA for guidance.] To compute the Rf values for each spot, use the “Rf Values Help” handout on the Linfield College chemistry website. The Rf for a compound is a constant value under given conditions, regardless of whether other compounds are present or not. If contaminants were present, they would likely show up as additional spots at different positions in a given lane (unless they have the same polarity as caffeine). CLEAN UP Recovered chloroform: Put in appropriate bottle in hood (“halogenated mixed organics”). Product: Throw away in trash. LABORATORY REPORT: Minor (skeleton) report Complete the electronic form for this experiment. Title Source reference Introduction containing: o experiment objective(s) (what was the purpose of performing the experiment?); o context (the experiment that was performed as a means through which the purpose could be met – pertinent concepts and techniques included); o rationale (why is this objective worthwhile?); and o Figure 1 should show only the molecular structure of caffeine with appropriate caption. Use ChemDraw to create the figure, highlight with the lasso tool, copy (Ctrl C), and insert in Fig. 1. Experimental methods (refer to the last page for examples of writing a concise experimental protocol) Results presenting actual data/observations that support whether the goal was met or not o Sentence(s) introducing values for % recovery and Rf, and introducing Figure 2 o Figure 2: Include a photo of your thin layer chromatography strip with an appropriate caption Discussion containing: o a reiteration of the objective, o interpretations of % recovery and Rf; what conclusions can be drawn about how much caffeine you were able to recover and how pure it seems to be, o plausible reasons for percent recovery deviations from 100 percent recovery or high purity, o chemistry concepts supporting results/interpretations, and o a summary sentence indicating whether or not objectives were met References (additional to source reference, and provide appropriate citations throughout the report) Appendix containing: o calculations with work for % recovery and Rf value o the nutrition label claims 2 g tea contains 55 mg caffeine – use this conversion for % recovery Note: Remember to cite sources in American Chemical Society (ACS) format. The direct link on our website which can help with proper citation format is: http://www.linfield.edu/assets/files/chem/Courses/CHEM%20321/ACSQuickRefGuide.pdf Additional help can be found using this online reference generator (and it’s also an app): https://www.refme.com You will also need to cite the references where you found the physical properties for your pre-lab materials, along with any additional external sources utilized as you complete your lab assignment. Enrichment questions (include answers in Appendix): 1. Was your product pure? How can you tell (what would you see if impurities - other compounds - were present)? 2. Did the substance you isolated from tea leaves behave like caffeine? Does the information you now have prove that you recovered caffeine and not some other compound? Why or why not? 3. Suppose that TLC revealed a contaminating substance along with caffeine. a. b. How would you know if there was a contaminant? How could you decide if it is more polar or less polar than caffeine? PRE-LAB QUESTIONS Week 3 only: 1. When you vacuum filter the tea leaves from the liquid after brewing the tea, where will you find the compound of interest – solids in the funnel or liquid in the flask? 2. During liquid-liquid extraction, is recovery of the compound more effective with a single extraction of a larger volume of organic phase, or several smaller volumes of the organic phase? Can you think of a reason for this? 3. What information from your Table 1 (materials chart) can help you predict which layer is on the bottom of a liquidliquid extraction? Compare to the same property for water, the aqueous phase. Week 4 only: 1. Identify what categories simple distillation, sublimation, thin-layer chromatography, and melting range are used for (isolation, synthesis, purification, and/or characterization). 2. What property does thin-layer chromatography take advantage of to characterize your product? 3. Explain the difference between % recovery versus % yield. Below are several experimental sections from published papers as examples to help you write yours. Note that each is complete, but concise. J. Org. Chem., 2011, 76 (2), pp 700–703 S. Chandrasekaran General procedure. Potassium carbonate (0.414 g, 3 mmol) was added to a solution containing the required bromosulfonium bromide (1 mmol) in CH2Cl2:H2O (1:1) mixture (20 mL). The corresponding activate methylene compound (2 mmol) was added to it and the reaction mixture was stirred for 8 h at room temperature. The CH2Cl2 layer was then separated and the aqueous layer was washed three times with dichloromethane (10 mL) and added to the organic layer. The combined organic layer was dried over anhydrous sodium sulfate and then evaporated. The residue was then purified by column chromatography on silica gel to give the corresponding doubly activated cyclopropanes in moderate to good yields. To a well-stirred solution of the corresponding cyclopropane derivative (1 mmol) in MeOH (4 mL) was added benzyltriethylammonium tetrathiomolybdate, 1 (0.731 g, 1.2 mmol). The reaction mixture was then stirred until the disappearance of the starting cyclopropane. The solvent was removed, the residue was extracted with CH2Cl2 (5 mL) and diethyl ether (20 mL) and filtered through a Celite pad. The residue was again extracted with CH2Cl2 (5 mL) followed by extraction with diethyl ether (20 mL) and filtered again through a Celite pad. The combined extract was evaporated and the residue was purified by column chromatography on silica gel to give the corresponding dihydrothiophene derivative J. Org. Chem., 2005, 70 (3), pp 1087–1088 Diethyl 1-(2-Azidoethyl)-2-oxopropylphosphonate (6). To a solution of tetrabutylammonium hydrogen sulfate (1.59 g, 4.68 mmol) in sodium hydroxide (2 M, 4.68 mL) was added a mixture of diethyl (2-oxopropyl)phosphonate (909 mg, 4.68 mL) and 2-iodoethyl azide (2.0 g, 10.3 mmol) in CH2Cl2 (4.68 mL). The resulting solution was refluxed for 36 h, cooled, and treated with water (20 mL) and CH 2Cl2 (20 mL). The organic layer was separated and concentrated under reduced pressure. The resulting residue was dissolved in Et 2O (100 mL) in order to precipitate tetrabutylammonium iodide. The salt was filtered off, and the filtrate was dried (Na 2SO4) and concentrated under reduced pressure to give a colorless oil. Flash chromatography on silica gel (3:1 hexanes/EtOAc) afforded 490 mg (40%) of 6 as a colorless oil. 3-(2-Azidoethyl)-3-eicosen-2-one (5). To phosphonate 6 (80 mg, 0.3 mmol) were added K2CO3 (700 mg), H2O (1.2 mL), tetrabutylammonium hydrogen sulfate (17 mg, 0.05 mmol), and heptadecanal (115 mg, 0.45 mmol). The resulting mixture was stirred at room temperature for 12 h, poured into water (10 mL), and extracted with CH 2Cl2 (3 x 15 mL). The organic layers were combined, dried (Na2SO4), and concentrated under reduced pressure. The residual oil was purified by flash chromatography on silica gel (100:1 hexanes/EtOAc) to give 87 mg (80%) of 5 as an inseparable 3.3:1 E/Z mixture J. Org. Chem., 1975, 40 (7), pp 966–967 (note: mols should also be given below for reagents) 2-(p-Bromophenylthio)furan (11). To a solution of n-butyllithium [prepared from n-butyl bromide (2.29 g) and lithium (0.297 g) in dry ether] was slowly added at -30oC 2-iodofuran 6 (3.68 g) in dry ether. The solution was allowed to reach ambient temperature and stirred for 2 hr. The reaction mixture was then cooled again at -70 oC and 4,4'-dibromodiphenyl disulfide (8.04 g) in dry ether was added. The reaction mixture was left overnight without further cooling, then hydrolyzed with HC1 (10%). From the ethereal layer, after concentration and vacuum distillation, was obtained 244-bromopheny1thio)furan (3.1 g), bp 120 oC (0.5 mmHg). J. Org. Chem., Vol. 62, No. 7, 1997 3,4-Bis(trimethylsilyl)thiophene (1a). To dihydrothiophene 9 (230 mg, 1 mmol) in chloroform (8 mL) was added a hot solution of DDQ (363 mg, 1.6 mmol) in chloroform (45 mL). The mixture was stirred at 60-65 °C for 10 h. The cooled reaction mixture was washed with aqueous sodium carbonate (10%, 3x-10 mL) and water (10 mL), dried (Na2SO4), and concentrated under reduced pressure. The residue was purified by column chromatography on silica gel (30 g, hexanes) to afford 1a (173 mg, 76%) as a colorless oil: J. Org. Chem., Vol. 65, No. 12, 2000 (how the solvent was remove should be given. Distillation? Reduced pressure?) 2-(Benzo[b]thienyl)-5-methoxybenzaldehyde (3). A mixture of 2-bromo-5-methoxybenzaldehyde (2) (1.5 g, 7.0 mmol) and Pd(PPh 3)4 (0.27 g, 0.73 mmol) in dimethoxyethane (40 mL) was stirred for 20 min in an argon atmosphere. To this mixture were added saturated NaHCO3 (10 mL) and benzo[b]thiophene-2-boronic acid (1) (1.36 g, 7.7 mmol), and the resulting mixture was refluxed with stirring. The progress of the reaction was monitored by TLC (5% EtOAc-hexane) for 1 h... The reaction mixture was cooled to RT, diluted with water (50 mL), and then extracted with CH2Cl2. The organic solution was separated, washed with 5% NaOH and water, and dried (Na2SO4). The solvent was removed to afford a solid which was recrystallized from EtOAc-hexane to yield 1.55 g (83%) of 3 as nearly colorless crystals, mp 125-127 °C. J . Org. Chem., Vol. 50, No. 22, 1985 N-tert -Butylthiophene-2-carboxamide (4). A mixture of thiophene-2-carboxylic acid (38.4 g, 0.3 mol) and thionyl chloride (71.4 g, 0.6 mol) was boiled under reflux for 3 h. Excess of thionyl chloride was removed by distillation under reduced pressure. The residue was taken up in CH2Cl2 (100 mL), and a solution of tert-butylamine (43.9 g, 0.6 mol) in CH2C12 (100 mL) was added with stirring at 10 oC. The resulting solution was stirred at 25 oC for 12 h, washed with water (3x, 30 mL), and dried (MgSO4). The combined washings were basified to pH 11 (concentrated aqueous KOH) and extracted with CH2C12 (3x- 30 mL) and the extracts dried (MgSO4). The combined organic solutions were evaporated under reduced pressure to give the crude product. Recrystallization (C6H6/CHCl3) gave the pure amide 4 (50.0 g, 91%) as a white solid: mp 144-145 oC *In general, please spell out chemical names in your reports rather than using formulas.