Introduction   Chemistry is an experimental science. In order to appreciate and comprehend the science of chemistry, one must actually participate in experimentation. To achieve good results from the practicals it is important that you: (i) understand the basic principles underlying each experiment, and (ii) should be able to discern which operations are critical and which are not.   It is therefore important that you READ THE PRACTICAL SHEETS before the class. If you do not understand any point please consult your instructor. A record of marks will be kept and will be used in assessment of progress at the end of the term. There will be a tutorial session (20 minutes) prior to experimental work . It is possible to obtain bonus marks through brilliance in the tutorial sessions. Marks obtained in this course count towards finals. During the lab period each student’s performance will be assessed subjectively. Where possible, students will work in pairs. Generally high marks will be awarded if the student goes about his duties with assurance, works safely, shows initiative and follows accepted scientific practices for lab work. Absence from the lab will mean loss of these performance marks. It is important to learn to use lab time efficiently. While waiting for a system to come to equilibrium, or for a solution to boil there will always be something constructive to do, such as preparing a graph, calculating preliminary data, or estimating new experimental values based on initial data points. The students effectiveness in utilizing such time will be considered when assessing the lab performance.       Basic instructions for laboratory work 1. Read the assignment before coming to the laboratory. 2. Work independently unless instructed to do otherwise. 3. Record your results directly onto your report sheet or notebook, DO NOT RECOPY FROM ANOTHER PIECE OF PAPER. 4. Work conscientiously to avoid accidents. 5. Excess liquid reagents and solutions should be disposed of by pouring them in the sink and washing them away with water; place excess solids in designated containers (not in the sink). NEVER RETURN REAGENTS TO THE REAGENT BOTTLE. 6. Do not place reagent-bottle stoppers on the desk; hold them in your hand. Your laboratory instructor will show you how to do this. Replace the stopper on the same bottle, never on a different one. 7. Leave reagent bottles on the shelf where you found them. 8. Use only the amount of reagent called for, avoid excesses. An undesired large amount of reagent might cause a lot of troubles during your tests. 9. Whenever instructed to use water in these experiments, use distilled water unless instructed to do otherwise. 10. Do not borrow apparatus from other desks. If you need extra equipment, obtain it from the stockroom. 11. When weighing, do not place chemicals directly on the balance. 12. Do not weigh hot or warm objects. Objects should be at room temperature. 13. Do not put hot objects on the desk top. Place them on a wire gauze or asbestos pad. 14. Do not heat measuring cylinders, burettes, pipettes or volumetric flasks on a direct flame. Please do not : 1. Smoke. 2. Eat or drink in the laboratory. 3. Leave clothes, unnecessary books etc., on the benches. 4. Leave apparatus in potentially dangerous places, i.e. near the edge of benches, or pipettes in tubes, bottles etc. 5. Pipette directly from distilled water bottles; decant into a beaker first. 6. Hesitate to ask. Some Common Laboratory Equipment and Glassware The CentrifugeEfficient centrifugation can only be obtained if the head of the centrifuge is well balanced . An unbalanced head will not only impair sedimentation of the particles in question, but may cause the spindle of the centrifuge to fracture with consequent destruction of the centrifuge and possible injury to bystanders. To avoid this, therefore :- 1. Always place tubes in the centrifuge in pairs, with the two members of the pair placed diametrically opposite to one another. 2. Opposite tubes must be of the same weight , and their contents must also be of equal weight. Contents may be balanced by eye, if the liquids in two tubes are of equal specific gravity (in larger, high speed centrifuges, the tubes must be weighed accurately on a sensitive balance). Check that the two centrifuge tubes are of equal length and wall thickness. 3. For centrifuges with swing-out rotors : the metal tubes in which the centrifuge tubes sit must be properly seated and free the swing outwards. Each tube must contain a rubber pad at the bottom. This is essential. If these pads are omitted, the glass tubes will fracture when subjected to high speeds and, in any case, the tubes will not be balanced. 4. The speed control knob must be turned fully anti-clockwise (i.e. to zero) before the centrifuge is switched on. Turn the speed up gradually. Do not turn speed up to maximum suddenly as this imposes a strain on the motor. 5. Do not open the centrifuge lid while the centrifuge is running. If a tube fractures , you could be showered with glass travelling at high speed ! 6. Leave centrifuge running for required length of time. Should any vibration or rattle develops, switch off at once, but do not open lid. Consult your instructor. 7. Turn speed control back to zero, and when the motor has stopped, open the lid.N.B. If a tube should break in the centrifuge, this will leave particles of glass in the tube pockets. These must be removed completely by thorough washing before the centrifuge is used again. Even microscopic particles of glass left in the pockets will cause other tubes placed in the centrifuge to break when the motor is started. Figure 1 An electric centrifuge Kipp’s Apparatus Kipp’s apparatus is a device used in laboratories for the production of a supply of any gas that can be evolved by the action of a liquid on a solid without heating. The simplest form is illustrated in Fig. 2. Figure 2 Kipp's apparatus   Opening the tap T allows the liquid in C to reach the solid in B. A reaction occurs and gas is produced. When the tap is closed, gas production continues until the liquid is forced back into C then A. Named after Petrus Jacobus Kipp (1808-1964). Fume Suction :If a fume hood is not available, substitute this apparatus.   Figure 3 A fume suction apparatus               Flame Test In flame test a 5 cm wire (preferably a platinum wire) is first cleaned by dipping in conc. HCl followed by heating the tip of the wire in a flame. When the flame acquires no colour, the wire is considered clean and ready for use. The tip of the clean wire is dipped in clean HCl then sprayed with the powdered sample and placed in the flame. The colour of the flame depends on the metal present in the material under investigation.   Figure 4 Platinum wire mounted in glass rod used in making flame tests for unknown ions.             The Glass ElectrodeGlass electrode is the most widely used electrode to measure the pH of a solution. Its operation involves the determination of electromotive force (e.m.f.) generated by a galvanic cell which may be represented as follows : In commerical pH meters, the glass electrode and reference half-cell are normaly combined in a single unit which can be dipped into the solution under test. A typical commercial combination electrode is shown in Fig. 5.                                                                             Figure 5 A commercial combination pH electrode The acidity of the test solution affects the overall e.m.f. of this cell at two points. One is the liquid junction between the saturated KCl and the test solution. The use of saturated KCl minimizes the liquid junction potential, and it is assumed that this potential is small and constant. The remaining contribution arises from the effect of the test solution on the potential across the glass membrane. The cell potential is given as E = constant + 0.0591 pH (at 25°C) The constant term incorporates the other interfaces in the cell and may be evaluated by immersing the electin a buffer solution of known pH. In practice the standardization is carried out by adjusting a control on the pH meter so that the pH of the buffer solution is registered correctly. The pHs of the standard and test solution need to be as close as possible for accurate results. The idea that glass membranes are semi-permeable to H+ ions is incorrect. This implies the change of H+ ions concentration inside the glass bulb (see Fig. 5). In fact, the membrane potentials appear because the silicate network of the glass has an affinity for certain cations e.g. H+ and Na+. These are adsorbed within the structure, and create a charge separation which alters the interfacial potential difference. At high pH values (low H+ ions concentration), the potential of a glass electrode becomes sensitive to Na+ concentration, and may register a pH of 12.8, for example, in a solution of pH 12.0. In practice, it is difficult to measure the pH very accurately using a pH meter. One reason for this is that the glass electrode tends to change its e.m.f. over a period of time. It is also affected by changes in temperature. However, for everyday use it is ideal. Because of restrictions set by the chemical composition of the usual glass membranes, glass electrode instruments may usually be used only in the range of pH values of pH 1 to pH 10. In actual practice, the following precautions will help to ensure the reliable measurement of pH. 1- The glass membrane must be kept clean. Electrodes should be immersed in distilled water or in buffer (pH 4 or 7) when not in use. 2- One must be careful not to allow films of sparingly soluble substances on the glass surface. 3- If accurate measurements in basic solutions are desired, pH measurements should be done in the absence of CO2, which is present in air. In any case the readings should be made quickly with minimal exposure to the atmosphere. 4- For solutions more basic than pH 11, glass electrodes of special composition are necessary to avoid interferences due to Na+. 5- To obtain stable operation of a new glass electrode, it is soaked in pH 7 buffer solution for at least 3 hours. If the electrode is allowed to become dry, it must be resoaked before being used again. 6- For pH measurements of vey dilute acids or bases (less than 0.01 M), addition of an electrolyte (such as KCl) may be necessary to provide sufficient conductance for stable operation of the pH meter. 7- Although glass electrodes have a plastic shield for protection, it is still possible to damage the tips if handled improperly. 8- A reference electrode usually has a hole for filling with electrolyte solution (e.g. 3M or saturated KCl solution). The reference electrode should be refilled with saturated or 3M KCl solution before getting dry. Care must be taken to prevent any of the sample being measured or any wash water from entering this hole. 9- Glass parts of the electrodes should not be touched with anything except soft tissue paper. 10- The following solutions are accepted as pH standards at 25°C. Standard Concentration pH Potassium hydrogen tartarate saturated 3.557 Potassium hydrogen citrate 0.05 m 3.776 Potassium hydrogen phthalate 0.05 m 4.008 Potassium dihydrogen phosphate 0.025 m 6.865 Sodium dihydrogen phosphate 0.025 m 6.865 Borax 0.01 m 9.180 Sodium carbonate 0.025 m 10.012 Note : m is molality (mol / kg) Reference : D. T. Sawyer, W. R. Heneman and J. M. Beebe, “Chemistry Experiments for Instrumental Methods” Johan Wiley & Sons, New York (1984) p. 17.                       Figure 6 Some typical laboratory equipment Figure 7 Individual water bath for heating reaction mixture           Figure 8 A capillary syringe for removing a supernatant liquid after centrifuging The Laboratory Notebook All students must use a lab book in which the lab preparation work and experimental results are recorded. Lab books will be evaluated regularly and at the end of the semester. Be sure to use your lab book to record everything, make notes, carry out calculations, etc. Do not be concerned about neatness (although this is an admirable trait), as much as you are about completeness. It is the responsibility of the student to have his lab book marked on a regular basis. To ensure that no one get too far behind in lab write-ups, students will not be permitted to do a new lab if they have more than one lab write-up incomplete. Your laboratory notebook should contain a permanent record of all your work in the laboratory and your thoughts about each experiment. It should be possible for you to write a complete description of any given experiment several months after you have conducted the laboratory work. During the experiment all data should be recorded in your laboratory notebook in ink. Do not feel constrained by the data table you have written. If extra data appear, write them down. Please do not trust your memory. Under no circumstances should you write data on spare scraps of paper. Take your notebook along. It is after all a book of notes on the experiment. If a piece of data is misrecorded, do not obliterate it like this: Instead, draw a single line through it: . You will often find that “bad” data were acceptable after all. Save all calculations for later ; record the raw numbers. For example, if you weigh a sample in a beaker and then subtract the weight of the beaker, record the weight of the sample plus beaker and the weight of the beaker. Leave a space in your data table for the weight of the sample. In this way, you will be able to avoid arithmetic mistakes that you might make during the laboratory period. You should also record all your tentative thoughts about the experiment. Your notebook should be a bound book, kept in ink with all pages numbered and dated. It should contain: 1. A table of contents. 2. A brief outline of the procedure. 3. Questions about the procedure and their answers. 4. Any deviations from the procedure as outlined. 5. All the raw data (without calculations) that were obtained . 6. Preliminary thoughts about the experiment. 7. All calculations performed to determine the final results . 8. The answers to all questions asked in the procedure and in the report.THE LABORATORY REPORT In this book, each experiment is followed by a REPORT FORM that should be carefully filled from the data gathered in your notebook. A good laboratory report is the essential final step in performing an experiment. It is in this way that you communicate what you have done and what you have discovered. Since it is the only means, in many instances, of reporting results, it is important that it be prepared properly. A laboratory report is a final draft. As such, it is always written in ink There should be no erasures or crossed out areas. The initial draft of laboratory report is in your laboratory notebook for two reasons: 1. It is unlikely that you will get everything correct on the first attempt and thus a first draft written on the report would be very messy. 2. If the report itself is lost or destroyed, you can easily and quickly rewrite the report from the notebook. It is realy essential that a laboratory report be neat. Studies have shown that when the same work is submitted in both neat and sloppy form , the neat version makes the better impression. This is true not only in academic work, but also in the business world . Neat work indicates that the writer knows the subject matter well. All data should be presented with the correct significant figures and units. Leaving off units makes it difficult for the reader to know the size of the numbers being reported. And writing down the wrong number of significant figures amounts to lying about the precision of the data. Too many significant figures implies that you know a number more precisely than you actually do. All questions should be answered with complete and grammatically correct sentences. Abbreviations should not be included in written answers . Read the sentence out loud to make sure that it makes sense. Your sample computations should be labeled with their purpose, for example, “mass of the liquid”. Within the computation, all numbers must have the correct units and the correct number of significant figures. Graphs Graphs are used to present the data in picture form so that they can be more readily grasped by the reader. Occasionally a graph is used to follow a trend. Notice that the best smooth curve is drawn through the data points. This is not the same as connecting the dots; note how the data points at the end of the curve in Fig. 9 (upper right) do not fall on the line. Usually, however, a graph is used to show how well data fit a straight line. The line drawn may either be visually estimated (“eyeballed”) or computed mathematically. There are many essential features of a good graph. 1. The axes must be both numbered and labeled. 2. The graph must have a title . 3. The data points are never graphed as little dots. One may use small circles, small circles with a dot inside, crosses, asterisks, or x’s. Dots are too easily lost on the graph or “created” by stray blobs of ink. 4. Any lines that appear on the graph in addition to data points should be explained. Thus, the line drawn is explained in the titles as “(visually estimated best straight line)”. 5. The scales of the axes should be adjusted so that the graph fills the page as much as possible. Treatment of Errors No measurement can be free of error and a good scientist must always try to assess the reliability of results. Errors can be broadly divided into two classes: (a) systematic errors which arise as a result of some bias in the apparatus (or the observer) and can be detected only by varying the experimental method; (b) random errors which arise due to slight, uncontrolled variations in the experiment and are assessed by comparing the results obtained from several runs, all made under the same conditions, so far as possible. The elimination of systematic errors generally constitutes a small research project in itself, but the treatment of random errors is a routine matter. Suppose measurements are made of a quantity x , yielding successive readings x1 , x2 ...... The average (algebraic mean) of n readings is defined by (1) The “true” result, , free of random errors, is taken as the limit of this average as the number of readings tends to infinity. Individual readings may be high or low. The best measure of the scatter is the “root mean square’ of the deviation, i.e. the standard deviation s, defined by (2) Again , we should take the limiting value of sn as n tends to infinity . For a so-called Gaussian or normal error distribution, 68% of all individual readings will lie within the range ± s. s is thus a useful measure of the probable random error in a measurement. The average of a finite number of readings will also show a small error, decreasing as the number of readings increases. A most useful result in statistical theory is that The probable error in an average <x>n is So by taking several readings we can make an estimate of , s and the probable error in our estimate of . A small sample of course cannot give the actual value of and we have to use <x>n instead ; the theory shows that the best estimate of s is then obtained from (3) A full statistical treatment fora sample of less that 10 readings is not really justified but it is an instructive exercise, nevertheless. In all cases, where 3 or 4 readings are obtained, you should make a rough estimate of the scatter and of the probable error in the average. You should make it a rule always to indicate the probable margin of error in any result, obtained either by comparing measurements or else simply by common sense (on balances, for example, the pointer wanders up and down on the scale or the digital numbers fluctuate around a certain value this gives an indication of the reliability of the result). Errors may be indicated either as absolute errors or as fractional or percentage errors e.g. (5.2 ± 0.1) cm or 5.2 cm ± 2% it is obviously not sensible to quote redundant figures : DON’T write (5.24333 ± 0.13333) cm It is equally wrong to omit trailing zero if they happen to be significant: DON’T write 2 cm if you mean (2.00 ± .01) cm. It is a standard convention in scientific work that, if no other indication of error is given, then the last figure quoted is subject to some error: thus 5 cm implies (5 ± 1) cm. but 5.00 cm implies (5.00 ± 0.01 ) cm. It often happens that two experimental results x, y have to be combined to get a final answer, z. In this case, the simple rules for calculating the error (dz) in the z are : (approximately) (a) if z = x ± y, dz = dx ± dy (b) if z = xy or x/y, (this includes the case z =1/y) Also, (c) if z = z = x/x. You should show the probable error at all relevant points in an extended calculation. Errors in calculation should not occur, but of course they will. The only safeguard is to think about the results as you go along. Particularly if you use a calculator, it is too easy to assume that a result “ must be right”. In fact it is easy to hit a wrong key and get a result which is widely wrong. Cultivate the trick of working the answer out in your head, just the first significant figure and the order of magnitude, as well as doing it on the calculator, as a check. Many experiments give a set of results which have to be plotted on a graph. In such cases two checks are readily made: are the results consistent and do they follow the expected curve? If there is a line going through several points, but missing another by a wide margin, this must be a strong indication of an error in calculation, or plotting. On the other hand, a smooth curve in place of a straight line may well indicate that you are using an incorrect method of calculation. When plotting graphs, give some thought to the choice of scale to make the best use of the paper. When the experimental errors are small (say, less than 1% of the range of values) straight line graphs should be plotted to give slopes of 30° to 60°, to give the best results for the gradient or for the intercept of the extrapolated line with an axis. If the experimental errors are large, it is important to mark in limits of error for each point, as illustrated in Fig. 9. Figure 9 Plotting a graph with errors in both x and y readings There is little point in expanding the scale beyond the stage where the error bars are more than 2 cm long, as plotted. In some cases this may mean that the graph looks like a horizontal line. This method of plotting makes it easier to estimate the possible error in the gradient or intercept, and to decide when the deviation of a point from the line is significant. To estimate the error for the gradient, draw “best” and “worst possible” lines through the points and measure the gradients on each. Errors in the intercepts can be obtained similarly. Some pocket calculators now provide “ least square” routines for calculating the “best fit”. These are useful but three points should be noted: (1) The least squares procedure used does not allow for possible errors in the x variable. (2) The probable errors are not calculated. (3) The calculator accepts precisely the numbers you give it, without criticism. It has no facility for spotting silly errors, which would be obvious on a graph. In general a graph gives a better “ feel” for the result, while the calculator may give a more precise answer. The methods are complementary: do not omit the graph!     Precision and Accuracy The only kind of physical quantity that can be determined with perfect accuracy is a tally of discrete objects, for example, books or the number of objects in a museum case. In measuring a quantity capable of continuous variation-for example, mass or length-there is always some uncertainty because the answer, like an irrational number such as p, cannot be expressed by finite number of digits. Besides errors resulting from mistakes made by the experimenter in the construction and use of measuring devices, other errors over which the experimenter has no control are inherent in measurements. Therefore, at least two, preferably three or more, determinations of any quantity should be made. The “true” value-more correctly the “accepted” value-of a quantity is chosen by some competent group such as a committee of experts as the most probable value from available data, examined critically for errors. The precision of a measurement is a measure of the mutual agreement of repeated determinations: it is a measure of the reproducibility of an experiment. The arithmetic average of the series is usually taken as the “best” value. The simplest measure of precision is the average deviation, calculated as follows: 1. Determine the average of the series of measurements. 2. Calculate the deviation of each individual measurement from the average. 3. Average the deviations (treat each as a positive quantity).Example: In a series of determinations, the following values for the molarity of a potassium permanganate solution were obtained: 0.1010, 0.1020, 0.1012, 0.1015 moles per liter. Calculate the average deviation. Average of the individual measurements Individual deviations from the average. Average: 0.1014 mol/l Average deviation: 0.0003 mol/l   These results would be reported as 0.1014 ± 0.0003 mol/l Frequently, precision is expressed as the relative average deviation, r.a.d., defined as the average deviation divided by the average. Thus the r.a.d. for the series of measurements in the above example is r.a.d. = (dimensionless) Multiplication by 100 yields the r.a.d. on a percentage basis. r.a.d. = ´ 100 = 0.3% If, for reasons of numerical convenience, the r.a.d. is to be expressed as “parts per thousand”, or “parts per million parts” (ppm), the fractional value may be increased by the appropriate multiplier. Thus, 0.003 (fractional) = 0.003 ´ 102 % or 0.3 % = 0.003 ´ 103 parts per thousand or 3 parts per thousand = 0.003 ´ 106 parts per million or 3 x 103 ppm The precision of an experiment varies with the method and apparatus used. Using the apparatus commonly available for quantitative analytical work, an experienced chemist can attain a precision of 1 part per 1000 or better in the gravimetric determination of the chloride in a water-soluble sample. The average inexperienced student more frequently obtains a precision of about 10 parts per 1000. With more complex analyses, the precision may decrease sharply. In planning an experiment, the experimenter should consider what overall precision is desirable and then choose the appropriate methods and equipment. Precise measurements, however, are not necessarily accurate. The accuracy expresses the agreement of the measurementwith the accepted value of the quantity. Accuracy is expressed in terms of the error (also called the absolute error), the experimentally determined value minus the accepted value. The relative error is the error divided by the accepted value. If the accepted value is unknown, the accuracy cannot be ascertained.Example. The accepted value for the molarity of the permanganate solution is 0.1024 mol/l. Calculate the error and relative error for the determination of the molarity in the previous sample.   from which the relative error is   ´ 100% = - 0.98 %Propagation of Errors When measured quantities are used to calculate another quantity, errors in the measurements introduce errors into the calculated result. The errors are said to be propagated through the calculations. When the error in each measured quantity has been estimated, the error in the result can be obtained, in simple cases, by the following rules. 1. The error in a sum or difference is the sum of the errors in the individual terms. Since the uncertainty is now in the third decimal place, we should round of the result to 1.893 ± 0.001g 2. The relative error in a product or quotient is the sum of the relative errors in the individual factors.Example (3) : Weight of object: 9.2152±0.0003 g Volume of object: 8.74 ± 0.07ml density = 0.8% + 0.003% @ 0.8% Hence, the error in the density = 0.008 ´ 1.05 @ 0.008. We may thus write the density as 1.05 ± 0.01 g/ml (0.008 may be rounded to 0.01). Significant Figures The number of significant figures in a quantity is the number of digits-other than the zeros that locate the decimal point-about which we have some knowledge. For example, the number of significant figures in 16.7193 is six; in 6.023 x 1023, four. When the last digit to be discarded is a 5, the number is rounded up or down in order to make the preceding digit even. For example. 6.85 ® 6.8 and 6.935 ® 6.94 When the error in a measurement has been estimated, the number of significant figures is immediately apparent. A digit that is uncertain by more than 6 or 7 should be discarded ; for example, 7.263 ± 0.006 covers the range 7.257 to 7.269, and therefore is better written as 7.26 ± 0.01 When numbers are used to calculate a result, the proper number of significant figures appearing in the answer can be decided by the following rules. 1. For addition and subtraction, the number of figures to the right of the decimal point in the sum (or difference) is equal to the number of figures to the right of the decimal point in the term that has the fewest such figures.Example: 0.784 + 15.16 - 9.6782 = 6.266 or 6.27 (rounded). There are two figures to the right of the decimal point in 15.16 and there should likewise be two in the answer. However, it is frequently desirable to carry one extra figure in calculations to minimize rounding error. Then round off the answer to the correct number of significant figures at the end of the calculation. 2. For multiplication and division, the number of significant figures (regardless of the position of the decimal point) in the product or quotient is equal to the number of significant figures in the factor that has the fewest significant figures. Example The answer contains three significant figures because that is the number of significant figures in 6.27. The other two factors have four each.   .