Research: Science and Education
How Do Organic Chemistry Students Understand and Apply Hydrogen Bonding? J. Henderleiter,*† R. Smart, J. Anderson, and O. Elian Department of Chemistry, Grand Valley State University, Allendale, MI 49401-9403; *
[email protected]
Hydrogen bonding is a basic chemical principle that has applications in all areas of chemistry. Chemistry students need to be able to analyze situations where hydrogen bonding can occur in order to understand reaction mechanisms, many physical properties, solubility, molecular interactions, and some spectroscopic information. This study examines how students completing a two-semester organic sequence understand, explain, and apply hydrogen bonding to determine the physical attributes of molecules. Background A working definition of hydrogen bonding at the undergraduate level can be found by evaluation of common texts. Some definitions appear very clear-cut; others are convoluted and may mislead students who have weaker chemistry backgrounds. Many introductory (general) chemistry texts define a hydrogen bond as a particularly strong dipole–dipole interaction between a hydrogen attached to an electronegative atom and an adjacent atom, ion, or molecule containing an electronegative atom (1–6 ). The electronegative atom, which has at least one lone pair, is often nitrogen, oxygen, or fluorine. Texts differ in that some prefer to introduce the bond as a moderately strong intermolecular force (3, 4, 6 ) instead of a type of dipole–dipole interaction (1, 2, 5). Some emphasize the polarity and partial charges of the atoms involved (1, 2, 4, 6 ). All use boiling-point trends as an example of a physical phenomenon influenced by hydrogen bonding. Viscosity, the density of ice, surface tension, and the structure of DNA are used as examples in which hydrogen bonding is also important (1–6 ). Organic texts define hydrogen bonding in different ways. Students with a weaker understanding of hydrogen bonding from general chemistry may have difficulty with the differences they encounter. For example, Volhardt and Schore (7) define hydrogen bonds as highly polarized bonds between hydrogen and an electronegative element. The proton-like character of the hydrogen allows it to interact strongly with anionic nucleophiles. In Fox and Whitesell (8), hydrogen bonds form between hydrogen and highly electronegative atoms from the fifth through seventh columns of the periodic table. This bond is the weak attraction of a hydrogen atom bonded to an electronegative atom for a lone pair of electrons on another electronegative atom. Jones (9) describes hydrogen bonding as it relates to the compounds of interest in various chapters. For example, hydrogen bonds are found in alcohols because the presence of oxygen atoms leads to highly polarized bonds and large dipole moments. Hydrogen bonding occurs when the basic oxygen atoms form partial bonds to acidic hydroxyl † Current address: Department of Chemistry, Western Michigan University, Kalamazoo, MI 49008-3842.
1126
hydrogens. Other definitions, such as that found in Solomons (10), are more similar to those found in general chemistry texts. Variations among texts make it difficult for students to connect prior knowledge from general chemistry to organic chemistry and may hinder students’ ability to analyze situations where hydrogen bonding influences physical properties. Previous Research Acquiring an understanding of a complex concept, such as hydrogen bonding, and being able to apply that concept to a variety of situations is not a trivial task. Learning a concept requires exposure to both positive and negative instances of the concept (11–13). Essential, nonessential, and irrelevant features of the concept must also be understood so that application and evaluation to novel situations is possible (11, 13, 14). As applied to hydrogen bonding, students must understand how covalent and ionic bonds differ from intermolecular forces. They must have an understanding of electronegativity as it relates to the distribution of electrons in molecules. The idea of bond polarity and how to determine bond polarity must also be present. At an introductory level, these concepts are traditionally explained as a set of rules about periodic trends and descriptions of atomic and molecular behavior, which itself is complex (15, 16 ). Students who possess misconceptions about atomic and molecular behavior or who simply memorize periodic trends may not be able to use their knowledge to explain physical properties such as boiling- and melting-point trends, solubilities of organic compounds, or certain characteristics of NMR and IR spectra. Many excellent laboratory activities and demonstrations require students to use hydrogen bonding and intermolecular force concepts to solve problems (17–20). However, little concrete evidence is provided to document how these activities improve student understanding. Peterson, Treagust, and Garnett (15) used a two-tier multiple-choice test to examine 11th- and 12th-grade Australian students’ understanding of covalent bonding and structure, including bond polarity, molecular polarity, and intermolecular forces. Bond polarity misconceptions included problems differentiating between electron pair position/sharing of electrons and ionic charge. Students misunderstood how molecular shape and bond polarity influence the polarity of molecules. Students also had trouble discriminating between intermolecular and intramolecular forces, and associated intermolecular forces with bonds in continuous covalent lattices. Schmidt (16 ) used a two-tier multiple-choice exam to study German senior high school students’ understanding of hydrogen bonds between organic molecules and their ability to predict boiling points of simple organic compounds. Students had trouble discriminating between molecules that could or could not hydrogen bond. They did not recognize the necessity of unpaired electrons and the necessity for hydrogen
Journal of Chemical Education • Vol. 78 No. 8 August 2001 • JChemEd.chem.wisc.edu
Research: Science and Education
Interview Questions 1.* Show where hydrogen bonds, if any, can form between any of the following molecules. H
H
H C H
H
C
H
H
H
Research Design
H O
H H
O
H
2. What other elements, if any, can you substitute for O, C, or H in the above molecules and still form hydrogen bonds? (Do not worry about the number of bonds that can form—we’re interested in what atoms and molecules can hydrogen bond.) 3. You overhear someone say that “hydrogen bonding only occurs between different types of molecules.” Is this true? Why or why not? 4. You overhear someone say that “hydrogen bonding only occurs between the same type of molecules.” Is this true? Why or why not?
5.* At the right is a molecule of bilirubin. Where, if anywhere, might intramolecular hydrogen bonding help stabilize this molecule? (Do not worry about the three-dimensional structure of this molecule—we’re interested in what parts of this molecule might be able to participate in intramolecular hydrogen bonding).
OH
O N
HO
N H H N
HO
N O
OH
6. You overhear someone say that “the reason that the boiling point increases from methanol to ethanol to propanol is because of hydrogen bonding.” Is this true? Why or why not? 7.* Which of the following compounds has the highest boiling point? O
O OH
OH
H
8.* The solubilities of ethanol and ethanoic acid (acetic acid) are roughly the same in water even though their boiling points are much different. Explain the reason for this observed solubility. O OH
OH
ethanol, bp 78 °C
ethanoic acid, bp 118 °C
9.* Intermolecular hydrogen bonding and intramolecular hydrogen bonding are affected differently by decreasing the concentration of the molecule being studied. Intermolecular hydrogen bonding effects decrease as concentration decreases, while intramolecular hydrogen bonding effects don’t change as concentration decreases. Why is this the case? 10.
Hydrogen bonding effects in NMR and IR spectroscopy are diminished as temperature increases. Why might this be true?
11.* Why is the O–H stretch in the IR so broad? [A textbook figure of a labeled IR spectrum of 1-hexanol, neat, salt plates, was provided to the interview subjects (21).] *Question given to the subject in written form.
to be directly bonded to an electronegative atom for hydrogen bonding to occur. Students’ ability to predict boiling points was hampered by their focus only on chain length but not molecular shape. Some also believed that boiling involves breaking covalent bonds.
On the basis of current literature (15, 16 ) and classroom experiences of two of us—JH and RS, who teach general and organic chemistry, respectively—we designed a series of 11 questions to probe students’ understanding of hydrogen bonding (see box). Questions were based on “typical” homework, laboratory, or exam questions that students encounter in organic chemistry courses. The questions also were designed to reflect the content taught, and presumably learned, from both general and organic chemistry. We hoped that students would be able to remember basic factual information about situations when hydrogen bonding could occur, and then apply that information to situations that might be novel. The text used in the organic course was Vollhardt and Schore’s Organic Chemistry, 2nd edition (Freeman, 1994); Bodner, Rickard, and Spencer’s Chemistry: Structure and Dynamics, preliminary edition (Wiley, 1996) was the general chemistry text used by many of the students participating in this study. A list of follow-up questions was constructed for use during the interviews to clarify student responses. The interviews were conducted by JH and JA, who read the questions to each subject. Where appropriate, figures and the written questions were available. Subjects were encouraged to write or draw during the interview, and all these materials were kept for analysis. The 20–30-minute interviews were audiotaped or videotaped, transcribed verbatim, and coded by JH and OE. The codes classified subjects’ statements into categories that demonstrated correct and incorrect conceptions of hydrogen bonding as related to definitions, occurrence, physical properties, and spectroscopic properties. Reasoning strategies used by students, for example reliance on rote memorization or the application of a definition, were noted. Interrater reliability was 85%. The Student Population Grand Valley is a public university concerned primarily with undergraduate education. Enrollment for fall 2000 was close to 18,500. The main campus is approximately 15 miles outside of Grand Rapids, Michigan. Of the students at Grand Valley, 90% are Caucasian, 4.4% African-American, 1.9% Hispanic, 1.7% Asian, 0.6% Native American, and 1.4% other or unclassified. The students come primarily from the three-county area surrounding the school (51.2%); 44.8% come from other parts of Michigan and 4% come from outside of Michigan. Sixty-one percent of the students are women and 39% are men. The average composite ACT scores for the middle 50% of students admitted in 2000 are 20–25. Interview subjects were volunteers completing the second course of a two-semester sequence of organic chemistry. Twenty-two students out of a class of 51 participated in the interviews. A comparison of the participants and the class as a whole is shown in Table 1. Participants tended to have a GPA higher than the class
JChemEd.chem.wisc.edu • Vol. 78 No. 8 August 2001 • Journal of Chemical Education
1127
Research: Science and Education Table 1. Demographics of Study Participants and All Students Completing 2nd-Semester Organic Chemistry, Winter 1998 Population
Variable
Study
Av. grade a
3.2
Entire Class 2.4
No. of Students Class:
Sophomore Junior Senior Non-degree seeking
Major:
Chemistry Biology Biological sciences/Pre-med Health Science Psychology Other/not specified
Gender: Female Male aOn
6 9 7 0
13 15 21 2
6 5 11 0 0 0
10 14 19 3 2 3
8 14
27 24
a 4-point scale, for two semesters of organic chemistry.
average. The gender mix does not match that of the class; proportionately more men participated in the study. The majors of participants are similar to those of the class as a whole. The sample is therefore not fully representative of students completing a year-long organic sequence; instead, it appears that academically stronger students participated in this study. Results Correct and incorrect conceptions about hydrogen bonding were classified into four categories: definitions or descriptions, occurrence of hydrogen bonding, physical properties, and spectroscopy. Correct and incorrect student conceptions for each category are discussed below. The responses we expected are also given.
Definitions and Descriptions Eighteen of the students interviewed provided an appropriate definition of hydrogen bonding, as determined from their answers to questions 1 and 2 in the box. We expected students to list nitrogen, oxygen, and fluorine as atoms capable of hydrogen bonding when hydrogen is bonded to one of these atoms in a molecule. The importance of lone pairs should have been noted. Students should have noted as well that hydrogen bonding could occur between the water molecules, but not between the methane molecules and not between the water and methane molecules. These 18 students accurately described the relationship between the electronegativity of the atom bonded to hydrogen and the resulting polarity of the bond. One student explicitly mentioned the need for lone pairs of electrons on the more electronegative atom. However, 13 students overgeneralized the atoms that are typically thought of as capable of hydrogen bonding (question 2), listing chlorine, sulfur, phosphorus, and carbon because of their proximity to nitrogen and oxygen (elements which students also stated were electronegative enough to be involved in hydrogen bonding). It appears that students relied on rote memorization to tell them which elements could be involved in hydrogen bonding. Though 1128
rote memorization of some facts is critical, in this case it seems that students memorized a list or a pattern but were not able to fully reason through it. Students did state that electronegativity increases across the periodic table from left to right, so elements to the right must be capable of hydrogen bonding. The strength or size of the dipole, presence of unpaired electrons, and relative sizes of the atoms involved were not addressed. The remaining four students confused hydrogen bonding with a covalent bond between hydrogen and some other atom. As the interviews with these students progressed, it became clear that the students’ original definitions of hydrogen bonding were of hydrogen bonds. These students did use hydrogen bonding as an intermolecular force to answer and explain the remaining questions.
Occurrence of Hydrogen Bonds Ideas about the occurrence of hydrogen bonding were revealed by responses to questions 3, 4, and 9. We expected answers to include a discussion of the ability of molecules that are capable of hydrogen bonding to do so with molecules of the same type and different types (e.g., water can hydrogen bond with water as well as with ammonia). Question 9 was designed to probe whether students understood or remembered the difference between intermolecular and intramolecular forces. All students correctly stated that hydrogen bonding could occur between molecules of the same type, such as water, and between different types of molecules, such as water and alcohols (questions 3 and 4). However, five students said that hydrogen bonds can be induced. They believed that the water molecules shown in the first question could induce the hydrogen atoms in an adjacent methane molecule to form hydrogen bonds with the oxygen in the water molecules. So molecules capable of hydrogen bonding, when adjacent to molecules that cannot hydrogen bond, cause the non-hydrogen-bonding molecules to form hydrogen bonds. Seventeen students reasonably predicted intramolecular hydrogen bonding. They discussed possible hydrogen bonding between hydroxy and amine groups in the bilirubin molecule shown in question 5. The remaining five students confused intramolecular hydrogen bonding with a chemical reaction. They explained that intramolecular hydrogen bonding within the bilirubin molecule resulted in formation of new covalent bonds through the creation of covalent bonds between the hydroxy groups and adjacent amines, or through condensation reactions between the hydroxy groups and amines. Physical Properties Students were able to predict trends in physical properties (questions 6–8). Though these questions are most appropriately answered by invoking intermolecular forces as well as energy considerations, molecular masses, and the like, the questions and answers focused exclusively on hydrogen bonding as a means of predicting trends. Our goal was to examine how students used hydrogen bonding to analyze the trends presented. We felt that this simplification would help focus the questions and answers more clearly on the phenomenon of interest. Sixteen students accurately related chain length or molecular mass to the boiling point trend observed for methanol, ethanol, and propanol (question 6). The remaining five could not explain the trend. One claimed to have memorized the trend (incorrectly), one tried to use pKa values to explain it, and the
Journal of Chemical Education • Vol. 78 No. 8 August 2001 • JChemEd.chem.wisc.edu
Research: Science and Education
remaining three claimed that resonance explained it. Three of these five students made statements suggesting they believed that boiling broke covalent bonds. Students had more trouble with question 7. Half of them correctly stated that more extensive hydrogen bonding explained why the carboxylic acid had a higher boiling point than an alcohol or aldehyde with a similar molecular mass. One student correctly predicted the pattern, but stated that the trend was memorized and offered no explanation for the pattern. Six other students also relied on memory, but their memory was faulty and they predicted that the alcohol would have the highest boiling point. Three students tried to use resonance arguments to predict the compound with the highest boiling point. The remaining student tried to use pKa values to determine the compound with the highest boiling point. Twelve students recognized the effect of hydrogen bonding on the solubility of organic compounds (question 8) and offered reasonable explanations for the similar solubilities of ethanol and ethanoic acid in water, related to their ability to extensively hydrogen bond with water. Four students offered no explanation for the solubility similarity—the boiling point differences confused them. The remaining six believed that chain length was the primary factor associated with solubility. When it was pointed out that the two compounds had the same number of carbons, these students could not come up with an explanation for the solubility data.
Spectroscopy Finally, students’ understanding of the influence of hydrogen bonding on spectroscopy was explored. Students at Grand Valley State University collect numerous IR spectra but only a small number of NMR spectra, owing to equipment availability. They are exposed to NMR and IR data through textbook examples and spectra provided by instructors in the laboratory. We were interested in exploring how students combine two topics that they may not previously have considered together. An appropriate answer to question 10, then, would include a statement that increasing the temperature would increase the speed of the molecules, and that this would likely weaken intermolecular forces. The weakening of forces would likely lead to changes in peak shape. Students were not asked or expected to predict the nature of these changes. Thirteen students explained why increasing the temperature might influence an NMR spectrum (question 10). They said that increasing the temperature would likely reduce the degree of hydrogen bonding, and this would somehow change the shape or size of the peaks. Four students had no conception of how temperature might matter. Five stated that heating or boiling the solvent or sample would break covalent bonds within these molecules. The three students who had stated this misconception earlier were in this group. Clarification of the question by telling students to assume the sample and solvent were not hot enough to boil did not help these students revise their answers. Question 11 was the most difficult question. Students were expected to state that hydrogen bonding would broaden the IR peak of 1-hexanol (neat, salt plates) because the stretching frequency of the OH bond would be influenced by hydrogen bonding to adjacent molecules. Twelve students offered no explanation whatsoever of why the broadening occurred. Of the eight students who gave a reason, two said that electro-
negativity differences were responsible for the broadening, four talked about the strength of the O–H bond, one talked about delocalization of electrons, and one said the broadening resulted from hydrogen bonding between alcohol and water (an impurity) in the sample. Two students correctly stated that hydrogen bonding was important to the peak broadening, but only one could explain how hydrogen bonding influenced the peak width. Recommendations These results illustrate that some students completing what is typically their second year of college-level chemistry still possess misconceptions found in younger, less experienced students. They have not abandoned—or have even formed— faulty beliefs, such as hydrogen bonds can be induced, intermolecular forces lead to reactions, or boiling breaks covalent bonds. These misconceptions make it difficult, if not impossible, for students to apply chemical concepts to data interpretation and analysis. Reliance on rote memorization as a means to analyze and interpret data is also problematic. Instructional strategies can and should be tailored to help students modify or eliminate their misconceptions, reduce their reliance on rote memorization for analysis and interpretation, and extend concept knowledge to novel situations. First, students must be guided to a better understanding of features that are essential, nonessential, and irrelevant to the concepts under study. Students must encounter examples and non-examples of important chemical concepts, in both the lecture and the laboratory as well as outside the classroom. Practice must be distributed over time during and between semesters. Students should encounter examples and nonexamples of concepts on multiple assessments. Assessments should, where possible, require students to describe why and how choices were made. Connections to critical concepts must also be made throughout multiple courses. This requires discussion among faculty so that those teaching at all levels have a reasonable understanding of what was and was not covered in earlier courses. These are time-consuming propositions, but if the goals of chemistry instruction include fostering critical thinking skills and equipping students with the skills and strategies needed to solve problems in contexts beyond those they were taught, then instructional practice must model and apply what is known about how people learn. Literature Cited 1. Bodner, G. M.; Pardue, H. L. Chemistry: An Experimental Science, 2nd ed.; Wiley: New York, 1995; pp 522–524, 678–679, 959. 2. Zumdahl, S. S.; Zumdahl, S. A. Chemistry, 5th ed.; Houghton Mifflin: Boston, 2000; pp 453–454, 1116–1117, 1128. 3. Ebbing, D. D.; Gammon, S. D. General Chemistry, 6th ed.; Houghton Mifflin: Boston, 2000; pp 460–464, 1003, 1080– 1082, 1090. 4. Brown, T. L.; LeMay, H. E. Jr.; Bursten, B. E. Chemistry: The Central Science, 8th ed.; Prentice Hall: Upper Saddle River, NJ, 2000; pp 399–403, 476–477, 1001. 5. Chang, R. Chemistry, 6th ed.; WCB McGraw-Hill: Boston, 1998; pp 422, 982, 988.
JChemEd.chem.wisc.edu • Vol. 78 No. 8 August 2001 • Journal of Chemical Education
1129
Information • Textbooks • Media • Resources 6. Jones, L.; Atkins, P. Chemistry: Molecules, Matter, and Change, 4th ed.; Freeman: New York, 2000; pp 426–429, 512, 514. 7. Vollhardt, P. K.; Schore, N. E. Organic Chemistry, 3rd ed.; Freeman: New York, 1999; p 195. 8. Fox, M. A.; Whitesell, J. K. Organic Chemistry, 2nd ed.; Jones & Bartlett: Sudbury, MA, 1997; pp 94–95. 9. Jones, M. Jr. Organic Chemistry, 2nd ed.; W. W. Norton: New York, 2000; p 767. 10. Solomons, G. Fundamentals of Organic Chemistry, 5th ed.; Wiley: New York, 1997; pp 77–78. 11. Ormrod, J. E. Human Learning: Principles, Theories, and Educational Applications; Merrill: New York, 1990; pp 304– 328. 12. Robert M. Gagne and M. David Merrill: In Conversation; Twitchell, D., Ed.; Educational Technology Publications: Englewood Cliffs, NJ, 1991.
1130
13. Tennyson, R. D.; Cocchiarella, M. J. Rev. Educ. Res. 1986, 56, 40. 14. Zoller, U.; Lubezky, A.; Naknleh, M. B.; Tessier, B. J. Chem. Educ. 1995, 72, 987. 15. Peterson, R. F.; Treagust, D. F.; Garnett, P. J. Res. Sci. Teach. 1989, 26, 301. 16. Schmidt, H. J. Presented at the Annual Meeting of the National Association for Research in Science Teaching, St. Louis, MO, April 1996; Eric Document Retrieval Service No. ED396914. 17. Hessley, R. K. J. Chem. Educ. 2000, 77, 203. 18. Bruist, M. F.; Smith, W. L.; Mell, G. J. Chem. Educ. 1998, 75, 53. 19. Wedvik, J. C.; McManaman, C.; Anderson, J. S.; Carroll, M. K. J. Chem. Educ. 1998, 75, 885. 20. Frohlich, H. J. Chem. Educ. 1993, 70, A3. 21. Brown, W. H.; Foote, C. S. Organic Chemistry, 2nd ed.; Saunders: Orlando, FL, 1998; p 515.
Journal of Chemical Education • Vol. 78 No. 8 August 2001 • JChemEd.chem.wisc.edu
