ARTICLES

A Laboratory Exercise on Semiquantitative Analysis of Ions in Nutrient Solutions

David R. Hershey* and Gary W. Stutte

ABSTRACT NUTRIENT SOLUTIONS

Astudent laboratory exercise on the developmentof nutrient Nutrient deficiency solutions for N, P, K, Ca, Mg, S, deficiencyand toxicity symptomswas linked to analytical tech- Fe, Mn, Cu, Zn, B, and Mo were prepared according to niquesand ion chemistryby havingthe studentsdetermine which Hoagland and Arnon (1950) except that Fe was provided ions in the nutrientsolutions were missing or presentin excess. as Fe-EDTAat 90/~M Fe, and CaCI2 was used instead Qualitativeanalytical methodsfor CI, Ca, Mg,Na, K, Mn,S, of CaSO4for the minus-N solution. The complete solu- ammonium,and nitrate were madesemiquantitative by using tion was a Hoagland Solution Number 1. standards.Quantitative analytical techniquesfor B, Fe, P, and Various element toxicities were induced by adding one bicarbonatewere simplified. Mostreagents and samples were of the following: 460 #M B(OH)3, 90/aM MnCI2, in dropperbottles, allowingstudents to rapidlyrun the tests mMNaC1, or 20 mMNaHCO3. The toxic levels of B by addingthe propernumber of dropsof sampleand reagent(s) and Mn were ten times the normal Hoagland Solution to a test tube. Formost elements, a color changeor precipita- tion denoteda positive test. Mostreagents have stored well for Number1 level. 2.5 yr, minimizingreagent preparation after the 1st yr. Ammoniumtoxicity was induced by growing in a modified Hoagland Solution Number1. The solution for ammoniumtoxicity had CaCl2 substituted for LASS LABORATORYEXPERIMENTS in , the Ca(NO)3, K2SO4 substituted for KNO3,and 15 mMam- C soilless culture of plants, are popular for demon- monium as (NH4)2SO4. strating nutrient deficiency symptoms(Epstein, Extra volumes of all but the minus-Zn, -Cu, and -Mo 1986; Moore,1974). Generally, nutrient deficiency symp- solutions were prepared for student analysis. Analytical tom laboratories are mostly observational in nature with techniques for Zn, Cu, and Mo were not attempted be- students focusing on symptom development. Tissue or cause the low concentrations of these elements are either nutrient solution analysis to confirm the nutrient defi- difficult to detect with simple techniques or difficult-to- ciency is usually not done due to a lack of analytical handle reagents are needed. Zinc analysis requires organic equipment. However, visual nutrient deficiency symp- solvent extraction with carbon tetrachloride (Epstein, toms have limited usefulness as a diagnostic procedure. 1986), Cu analysis requires toxic cyanides, and Mo Introductory inorganic chemistry laboratories often analysis requires atomic absorption spectrophotometry perform qualitative analysis of ions in solution whenstu- (American Public Health Association, 1985). The all- dents learn ion chemistry in a hands-on setting. A weak- nitrate-N, all-ammonium-N, nine deficient, and four ness of this classic qualitative analysis laboratory toxic nutrient solutions were assigned numbersand placed experience is that the ion solutions analyzed, particularly in 30-mL-glass-dropper bottles. the unknowns,are artificial. In the real world, plant scien- tists knowwhat they are analyzing be it irrigation , solution, plant tissue, or soil. PLANTS In order to extend the learning experience in a plant Leaf cuttings of Tolmiea menziesii (Pursh) Torr. & nutrient deficiency laboratory and link it more closely to Gray were rooted in 2 mMCaC12 in solution culture analytical chemistry, qualitative analysis laboratory tech- (Hershey,1989), placed in plastic jars in continuouslyaer- niques were utilized to analyze nutrient solutions. The ob- ated nutrient solutions, and set on a greenhouse bench jective was to determine if an ion or element was present for symptom development. The aeration manifold was at deficient, adequate, or toxic levels. like that of Hershey and Merritt (1986) as modified Sadof and Hershey (1989). There were four replications Department of Horticulture, Univ. of Maryland, College Park, MD per treatment, but not all jars were labeled for the stu- 20742-5611. Received 7 May 1990. *Corresponding author. dents. The unknownjars were labeled for the instructors, Published in J. Agron. Educ. 20:7-10 (1991). and wereused later to test the student’s ability to discern

J. Agron. Educ., Vol. 20, no. 1, 1991 7 the identity of the deficient or toxic element via symptom Table 1. Directions for achieving desired molarity of reagents identification and solution analysis. The seed-propagated necessary for the nutrient solution analysis laboratory. The in- crops more often used in nutrient deficiency laboratories, structions for each test are given in the text. such as tomato (Lycopersicon esculentum Mill.) (Epstein, Reagent Preparation 1986) or sunflower (Helianthus annuus L.) (Moore, 1974) Chloride could also be used. The houseplant Tolmiea, knownas 17 g AgNO3L-1 0.1 .~/AgNO3 -1 piggyback plant, was chosen for a change of pace and 10 mMNaC1 584 mg NaCI L because its rosette form requires no staking. Sulfate 1 MBaC12 244-1 g BaC12.2H20 L 1 mMH2SO4 56-1 #Lconc. H2SO4L ANALYTICAL PROCEDURES Phosphate Reagent A 12 g ammoniummolybdate [(NH4)6Mo7024-4H20]in 250 mLH20 and Qualitative analysis procedures for Cl, Ca, Mg, Na, 0.3 g antimony potassium tartrate K, Mn, ammonium,nitrate, and sulfate (Masterton and (CaH4K2Sb2012’3H20)in 100 mLH20 are Slowinski, 1978) were used with modifications. Standard added to 1 L of 2.5 M H2SO41148 mL conc. H2SO4 L-1L mixed well, and qualitative analysis procedures usually utilize samples brought to 2 L with 20 mMconcentrations of each ion. This presented Reagent B 0.25 g ascorbic acid in 50 mLof Reagent A Imake fresh) sensitivity problemsfor certain ions since a HoaglandSo- 1 mMKH2PO4 136-1 mg KH2PO4L lution Number1 has ion concentrations of 15 mMorless Nitrate (Hoagland and Arnon, 1950). To overcome the problem, 18 M H2SO4 conc. H2SO4 solutions for the least sensitive tests, Mnand Mg, were 0.2 MFeSO4 2.8 g FeSO4-7H20in 50 mL(make fresh) 10 mM KNO3 1.01 g KNO3L-1 evaporated to dryness in an oven and reconstituted to Ca one-tenth of the original volume prior to class. With the 6 M NH40H 400-1 mL 30% NH4OHL 1 Mpotassium oxalate 184-1 g K2C204.H20 L exception of C1, the other tests were sensitive enoughto -1 detect deficient or adequate ion levels. The C1 test was 5 mMCaCl2 735 mg CaC12.2H20 L Mg only used to detect a Cl excess. 6 M HC1 500-1 mLconc. HC1L Quantitative analytical procedures for B (John et al., Mgreagent 0.1g p-nitrobenzene-azoresorcinol L-1 of 25 mMNaOH I1 ~ L -11 1975, bicarbonate (Chapmanand Pratt, 1961), phosphate 6 M NaOH 240 g NaOH L-~ (Watanabeand Olsen, 1965), and Fe (Chaney et al., 1972) 5 ram MgSO4 1.23-1 g MgSO~.7H20L were simplified by minimizing the solution volumes uti- Ammonium 6 M NaOH 240 g NaOH L-1 lized, measuring reagent volumes by drop-count instead 10 mMNH4C1 535-1 mg NH4C1L of pipettes, and visually estimating solution color instead K of using spectrophotometers. 10 mMKC1 746 mg KC1L- 1 For the majority of tests, 10 drops of sample are added Na to a 12- by 75-mm-glass-test tube and the appropriate 10 mMNaCl 584 mg NaCI L-1 Mn number of drops of one or more reagents added. A posi- 6 M HNO3 400-1 mL conc. HNO3L tive test is indicated by a changein solution color or for- Sodium bismuthate -1Solid NaBiO3 mation of a precipitate. Sections of 512 square, plastic 100 ~ MnCl2 20 mg MnCI2.4H20 L plug flats, used for seed germination in bedding plant Buffer masking reagent 125 g ammoniumacetate (CH3COONH4) production, are utilized as inexpensive racks for the 12- and 7.5 g Na-EDTAin 200 mLH20; by 75-mm-disposableotest tubes. Each plug cell is 14 mm slowly add 62.5 mLglacial acetic acid Azomethine-H reagent 0.45 g Azomethine-Hand 1 g ascorbic acid wide and 21 mmdeep. The flats are available from in 50 mLH20 (make fresh) Maryland Plants and Supply, Baltimore, MD.Most rea- 500 ~/B{OH)a 31 mg B(OHh -1 gents and nutrient solutions are in 30-mL-dropperbottles. Fe Standard solutions of specified concentrations are avail- 2.5 mMTPTZ (2,4,6 tri- pyridyl-s-triazine) 780 mg TPTZ L-1 able for each ion and allow the student to judge if the 4 M sodium acetate 544-1 g CHaCOONa-3H20L nutrient solution concentration is higher or lower than 2 M HC1 165 mL conc. HCI L-1 Ascorbic acid Solid the standard. Deionizedwater is available in dropper bot- 100 ~ Fe-EDTA 36.7-1 mg Fe-EDTA L tles so students can determine what a negative test result Bicarbonate looks like. Preparation of reagents for each test are sum- Phenolpthalein indicator 2.5 g phenolpthalein L-1 of 1:1 ethanol~water marized in Table 1. Details for each test are described Methyl orange indicator 1 g methyl orange L- 1 below. 25 mMH2SO4 1.39-~ mLconc. H2SO~L -1 Chloride. To 10 drops of sample add two drops of 0.1 10 mM NaHCO3 840 mg NaHCOaL M AgNO3.A cloudy white precipitate of AgCl indicates the presence of chloride. The dropper bottle of AgNO3 solution is wrapped in aluminum foil to prevent pho- Sulfate. To 10 drops of sample add three drops of 1 todegradation of AgNO3. A 10-mM-NaCl solution is MBaCl2 and mix well. A white, finely divided precipi- used as a standard. This test is not sensitive enough to tate of BaSO4 indicates the presence of sulfate. A 1-mM- detect the normal 18/d~/-chloride concentration in H2SO4 standard is used. Hoagland Solution Number 1 Hoagland Solution Number1, so it is used to detect an has 2 mMsulfate. excess of chloride. Phosphate. To one drop of sample add four drops of

8 J. Agron.Educ., Vol. 20, no. 1, 1991 Reagent B (Table 1) and 20 drops of deionized water. Mix flame. Dip the wire loop in the sample and place in the and let stand for 10 min. A blue solution indicates the flame. A violet flame, lasting 0.5 s indicates the presence presence of phosphate. A 1-mM-KH2PO4standard is of K. used. Hoagland Solution Number 1 contains 1 mM If the sample solution contains Na, the violet K flame phosphate. will be obscured by the stronger yellow Na flame. If this Nitrate. Add10 drops of sample to a test tube. Care- is the case, view the flame through one or two layers of blue glass or blue plastic film, whichfilter out the fully add 20 drops of 18 M H2SO4. Add the acid slow- ly, mixing continuously. Be careful because the tube yellow color due to Na and allow the violet flame to be becomes hot. Cool the tube under a water tap. Holding observed. the tube at a 45 ° angle, let five drops of 0.2 MFeSO4 The Hoagland Solution Number 1 has 6 mMK. Ten run downthe side of the tube and form a layer over the millimolar KCI is the standard. acid. The FeSO4should be prepared fresh for each lab. Sodium. Run the Na flame test as for K. A strong yel- Let the tube stand for a few minutes. A brown ring at low flame, persisting for 1 to 2 s confirms the presence the junction of the two layers indicates the presence of of Na. Ten millimolar NaCl is used as a standard. nitrate. Ten millimolar KNO3serves as a standard. Manganese.Put 10 drops of sample solution in a test . Put 10 drops of sample in a test tube, then tube, and add 10 drops of 6 M HNO3.With a stainless add five drops of 6 M NH4OHand mix. Add five drops steel spatula, add about 0.1 g of solid NaBiO3(sodium of 1 M K2C204(potassium oxalate), mix and let stand bismuthate) and mix. There should be a little solid bis- for 1 min. A white precipitate of calcium oxalate is con- muthate in excess. Let the mixture stand for a few firmation for the presence of Ca. Hoagland Solution minutes. A purple solution confirms the presence of Mn Number 1 has 5 mMCa, so 5 mMCaC12 is used as a as permanganate. Sodiumbismuthate oxidizes Mnto per- standard. manganate. Magnesium.Put 12 drops of sample in a test tube, add This test is not sensitive enough to detect the 9 ~ Mn three drops of 6 M HC1, and three drops of magnesium in Hoagland Solution Number1. Therefore, samples are reagent (Table 1). The solution will turn yellow. Stir the concentrated tenfold, e.g., 100 mL of sample is evapo- solution and add, drop by drop, 6 MNaOHuntil the so- rated to dryness and dissolved in 5 mLof deionized water lution turns purple or a blue precipitate forms. A and 5 mL of 6 M HNO3.Twenty drops of this concen- medium-blue precipitate of magnesium hydroxide with trated sample is used in place of the 10 drops of sample adsorbed magnesiumreagent indicates that Mgis present. and 10 drops of HNO3.A 100 #M MnCI2 solution is The Mgtest is one of the least sensitive, so the 2 mM used as a standard. Mg in Hoagland Solution Number 1 is too low to give Boron. Put 10 drops of sample in a plastic, dispos- a positive test. To successfully use the test, the sample able, 1-mLcuvette. Borosilicate glass test tubes should solutions are concentrated tenfold, e.g., 100 mLof sam- not be used because they may produce a false positive ple is evaporated to dryness in an oven and dissolved in reaction. Add 8 drops of buffer masking reagent and 2 8 mL deionized water and 2 mL of 6 M HCI. Fifteen drops of azomethine-H reagent (Table 1). Mix thorough- drops of this concentrated solution are used in place of ly, and allow the color to develop at room temperature the 12 drops of sample and three drops of 6 M HCI. for at least 30 min. A yellow solution indicates the Use 5 mMMgSO4 as a standard since it is the mini- presence of B. mumdetectable concentration with this test. Prepare azomethine-Hreagent just before lab and store Ammonium.Put 10 drops of sample into a 50 mLplas- in a refrigerator. tic beaker. Moisten a piece of red litmus paper with Hoagland Solution Number1 contains 46 ~ B as un- deionized water and stick it on the convexside of a small dissociated boric acid, B(OH)3.A 500 talk/B(OH)3 stan- watch glass. Add 10 drops of 6 M NaOHto the beaker dard is used to demonstrate toxic concentrations. For an and gently swirl to stir. Cover the beaker with the watch optimal level standard, use one drop of 500 ~ B(OH)3 glass, convex side down. The litmus paper gradually turns and 9 drops of deionized water in place of the 10 drops blue if ammoniumis present due to formation of am- of sample. monia gas. Care must be taken that no solution splashes Iron. Add 20 drops of sample to a test tube followed on the litmus paper and gives a false positive. The watch by one drop of 2.5 mMTPTZ(2,4,6-tripyridyl-s-triazine), glass can be removed after the litmus paper turns blue, two drops of 4 M sodium acetate, and two drops of 2 and you may be able to detect the odor of ammonia. M HC1. Add a pinch of ascorbic acid using a stainless Hoagland Solution Number 1 has no ammonium. The steel spatula and mix. A blue solution indicates the all-ammonium solution has 15 mMammonium. Ten mil- presence of Fe. The ascorbic acid reduces the ferric Fe limolar NH4C1is used as a standard. in Fe-EDTAto ferrous Fe which forms a chelate with Potassium. Potassium and Na are determined by flame TPTZ. emission tests. A bunsen burner or inexpensive, porta- The Hoagland Solution Number 1 was modified to ble propane torch is needed along with a platinum wire contain 90/zMFe as Fe-EDTA. The standard is 100 ~ loop. Fe-EDTA. Dip the end of the platinum wire in 12 M HCI, and Bicarbonate. This test is usually used for water analy- flame the tip of the wire until it imparts no color to the sis where carbonate and bicarbonate are the major ions

J. Agron. Educ., Vol. 20, no. 1, 1991 9 which buffer the solution pH. For this lab, the nutrient (Gleditsia triacanthos var. inermis Willd. 'Sunburst') were solutions contain appreciable orthophosphates, which be- placed in a Hoagland Solution Number 1. Students typi- have like bicarbonate in the titration. Thus, this test does cally guess that the solution is deficient in Fe because the not confirm the presence of bicarbonate. However, high youngest leaves are bright yellow, but solution analysis readings from this test are likely to be due to bicarbonate, does not confirm the diagnosis. Sunburst honeylocust was rather than phosphate. selected by horticulturists for the unusual appearance of Using a graduated cylinder, put 25 mL of sample into the new leaves (Lancaster, 1974). The yellow leaves are a 125-mL-Erlenmeyer flask. Add two drops of phenol- normal, not an Fe deficiency. There are thousands of cul- phthalein indicator (Table 1). tivars of variegated plants which could be used instead If the solution turns pink, titrate with 25 mM H2SO4 (Yokoi and Hirose, 1978). Other potential trick plants, until colorless. The titration is done with a burette sup- whose symptoms appear to be caused by nutrient- ported by a ringstand or by counting the drops of acid deficient solutions, but are not, include plants grown with added using a dropper. Record the milliliters of acid used low root temperatures to induce deficiency, and designate as y. To convert the number of drops to plants overdosed by herbicides which suffer from chlo- milliliters, divide by the appropriate number determined rosis, or insect-damaged plants. by adding drops to a 10-mL-graduated cylinder. Next add two drops of methyl orange indicator (Table 1). Titrate to the orange endpoint, again recording the milliliters of acid added, and designating it as z. Approximate the millimoles of carbonate as y and the millimoles of bicarbonate as 2z. Normally, y will be zero. A 10 mM NaHCO3 standard is used. CLASS APPLICATION

The analysis lab was conducted the week after the so- lution culture study was started. Each student was given a data sheet listing the 13 tests and indicating which solutions were to be tested. Deionized water, an appropriate standard for the ion or element, Hoagland Solution Number 1, and the solutions deficient and toxic for that ion or element were tested. Therefore, there were four or five solutions for each test. Equipment and samples for each test were at a single location, so one or two students could complete the test for one element and then move on to the next. After the instructor demon- strated to the class how each test was performed, a class of 15 students took about 1 h to complete the tests. Laboratory safety was stressed since concentrated acids or bases are used in many of the tests (Nagel, 1989). After deficiency/toxicity symptoms on solution culture plants were well developed, students were asked to iden- tify what was wrong with the unknown plants and then confirm their diagnosis by analyzing the nutrient solu- tion. A few trick plants were included to test students' knowledge of the limitations of nutrient deficiency symp- toms. For example, new shoots of Sunburst honeylocust

10 J. Agron. Educ., Vol. 20, no. 1, 1991