Alternative Conceptions Of Chemical Bonding

Research in Science & Technological Education, Vol. 19, No. 2, 2001

Alternative Conceptions of Chemical Bonding Held by Upper Secondary and Tertiary Students RICHARD K. COLL, School of Science & Technology, The University of Waikato, New Zealand

NEIL TAYLOR, School of Education, The University of Leicester, UK

ABSTRACT Examination of senior secondary and tertiary level chemistry students’ descriptions of their mental models for chemical bonding revealed prevalent alternative conceptions. In addition to some common alternative conceptions previously reported in the literature, such as misunderstandings about intermolecular forces and molecularity of continuous lattices, the inquiry found a surprising number of alternative conceptions about simple ideas like ion size and shape. Some 20 alternative conceptions were revealed, the most common being belief that continuous ionic or metallic lattices were molecular in nature, and confusion over ionic size and charge. It is posited that the mass of curriculum material students encounter during their undergraduate and postgraduate studies may have some inuence on the formation of alternative conceptions. Hence, it is recommended that tertiary level teachers in particular consider the advisability of limiting the teaching of some abstract models for chemical bonding until an advanced stage of the undergraduate degree.

Introduction There has been something of a revolution in science education since the 1960s. Changes in the way science education researchers, philosophers and others see the world has resulted in a major rethinking of how science should be taught. Prior to the 1970s, science teaching was dominated by a transmissive approach. Implicit in this approach was a view that the learning of science was passive and that knowledge could be ‘piped’ from the full container of the teacher’s head to the empty vessel of the student’s head (Tobin et al., 1990). Little account was given to students’ alternative conceptions of science, which were considered to be easily extinguished or replaced by the teacher through persuasive argument. This resulted, in general, in a very didactic approach to the teaching and learning of science. However, in the 1970s new cognitive theories began to emerge which challenged this passive view of learning. An emphasis was placed upon the student as an active individual reaching out to make sense of events and constructing knowledge through social interaction and experiences with the physical environment (Driver & Easley, 1978). These new theories acknowledged that, contrary to the view that students have blank minds, they bring to their school learning in science ideas, expectations and beliefs concerning natural phenomena which they have developed to make sense of their own ISSN 0263-514 3 print; 1470-113 8 online/01/010171-2 1 Ó DOI: 10.1080/0263514012005771 3

2001 Taylor & Francis Ltd

172

R. K. Coll & N. Taylor

past experiences. Furthermore, these ideas could differ from the currently accepted scientiŽ c view, and from the intended learning outcome, and could be extremely resistant to change (Driver, 1981). Such thinking led to the development of the constructivist paradigm which has as a basic premise that knowledge is created in the mind of the individual rather than absorbed or transmitted from an expert or teacher to a student (Driver, 1989). The constructivist view of knowledge acquisition led to changes in the nature of teaching and research. Much research into science teaching revealed that students hold many views that are at variance with commonly accepted scientiŽ c views. In fact so proliŽ c has this research been that there are now substantial bibliographies, with over 1000 references, into investigations of students’ conceptions in science (e.g. Pfundt & Duit, 1997). Despite the proliferation of studies into students’ understanding of various aspects of science, there has been relatively little investigation of their understanding of chemical bonding, particularly amongst senior secondary and tertiary level students. In this article we address this issue by identifying a range of alternative mental constructs of chemical bonding and consider their origins and possible implications for teaching at this level. DifŽculties in the Teaching of Abstract Chemistry Concepts There seems to be a widespread perception amongst researchers and teachers that many students Ž nd chemistry difŽ cult (Carter & Brickhouse, 1987; Nakhleh, 1992; Barrow, 1994; Kirkwood & Symington, 1996). The reason suggested is that chemistry is a complex subject possessing many abstract, frequently counter-intuitive concepts (Gabel, 1998). Furthermore, Hawkes (1996) and others (e.g. Fensham & Kass, 1988; Taber, 1995a) point out that there are many alternative conceptions in commonly used chemistry textbooks. Remarkably, Hawkes stated that ‘after writing an article on textbook errors I received a letter from a Nobel Laureate expressing disbelief in my statement that only 2% of aqueous CdI2 exists as Co2 1 (aq.)’ (p. 421). One of the essential characteristics of chemistry is the constant interplay between the macroscopic and microscopic levels of thought, and it is this aspect of chemistry (and physics) learning that represents a signiŽ cant challenge to novices (Bradley & Brand, 1985). Numerous reports support the view that the interplay between macroscopic and microscopic worlds is a source of difŽ culty for many chemistry students. Examples include the mole concept (Gilbert & Watts, 1983), atomic structure (Zoller, 1990; Harrison & Treagust, 1996), kinetic theory (Abraham et al., 1992; Stavy, 1995; Taylor & Coll, 1997), thermodynamics (Abraham et al., 1992), electrochemistry (Garnett & Treagust, 1992; Sanger & Greenbowe, 1997), chemical change and reactivity (Zoller, 1990; Abraham et al., 1992), balancing redox equations (Zoller, 1990) and stereochemistry (Zoller, 1990). Student Alternative Conceptions of Chemical Bonding The research that exists on students’ view of chemical bonding has revealed prevalent and consistent alternative conceptions across a range of ages and cultural settings. Work on the understanding of intermolecular bonding has provided some evidence that students appreciate the relationship between intermolecular bonding and physical properties such as boiling point (e.g. Peterson & Treagust, 1989; Peterson et al., 1989; Taber, 1995b, 1998; De Posada, 1997; Taylor, & Lucas, 1997). However, other research reveals that students believe intermolecular bonding is stronger than intramolecular bonding (Peterson et al., 1989; Goh et al., 1993), and that they invoke intramolecular

Alternative Conceptions of Chemical Bonding Held by Students

173

bonding in inappropriate circumstances (e.g. in ionic compounds) (Taber, 1995b, 1998), or believe it is absent in polar molecular substances such as water (GrifŽ ths & Preston, 1989; Birk & Kurtz, 1999). A highly prevalent alternative conception for chemical bonding is that continuous covalent or ionic lattices contain molecular species (De Posada, 1997; Taber, 1998; Birk & Kurtz 1999). Butts and Smith (1987) suggest that the ubiquitous use of ball-and-stick models used to model ionic lattices may be instrumental in the generation of this alternative conception because students mistake sticks for individual chemical bonds. The fact that other research revealed that students believed ionic substances such as sodium chloride possessed covalent bonds adds credence to this suggestion (Peterson et al., 1989; Taber, 1994, 1997). A related alternative conception, reported by Boo (1998), is that some students believe that a chemical bond is a physical entity. Boo suggests that this arises from a worldview that building a structure requires energy input, whereas destruction involves release of energy—that is, students believed that bond breaking releases energy and bond making involves energy input. Confusion about the concept of electronegativity is also widespread, resulting in a number of alternative conceptions for chemical bonding; inability to establish the correct polarity of polar covalent bonds, the view that non-polar molecules are only formed between atoms of similar electronegativity and that the number of valence electrons, the presence of lone pairs of electrons, or ionic charge determine molecular polarity (Peterson et al., 1989; Harrison & Treagust, 1996; Boo 1998; Birk & Kurtz, 1999). Students appear to have little appreciation of the underlying electrostatic nature of chemical bonding (Taber, 1995b; De Posada, 1997; Boo, 1998). For example, attraction between two oppositely charged species was thought to result in neutralisation rather than bond formation, the likely source of confusion being the parallel with acid–base chemistry (Schmidt, 1997; Boo, 1998). Similarly, students have a poor understanding of the bonding in metals, seeing metallic bonding as unimportant or in some way inferior to other forms of bonding, despite being able to use the common sea of electrons model to explain the properties in metals (Taber, 1995a, 1998; De Posada, 1997). There are a number of other alternative conceptions about covalent bonding reported in the literature. Some students believe that the number of valence electrons and the number of covalent bonds are one and the same; other conceptions include confusing resonance forms with molecular structures and believing that covalent bond formation involves the transfer of electrons (Taber, 1994, 1997, 1998). Much of the research cited above was undertaken with students of school age. The conceptions which tertiary level chemistry students hold about chemical bonding have, in comparison, been researched infrequently. Perhaps by this stage in their academic careers, students may be perceived as having a sound understanding of scientiŽ c concepts, and are therefore less prone to developing nai¨ve mental models. In this article we describe a study of senior secondary school and tertiary level chemistry students from New Zealand and Australia. The study revealed a number of alternative conceptions for chemical bonding and these, along with the implications of the study for the teaching and learning of senior level chemistry are discussed. Methodology Theoretical Framework The work reported here represents part of a larger study into students’ mental models for chemical bonding (Coll & Treagust, in press). This inquiry is a naturalistic inquiry

174

R. K. Coll & N. Taylor

conducted within a constructivist paradigm. The authors believe inquiries into science education such as that described in this work are best addressed using a methodology in which the individual constructions of interest are elicited by interactive dialogue between researchers and participants. SpeciŽ cally the authors subscribe to a social and contextual constructivist belief system. Contextual constructivists assert that a crucial feature of knowledge creation is that it is not carried out in isolation, but is subject to in uence by an individual’s context; that is, his or her prior knowledge and experiential world (Wheatley, 1991; Cobern, 1993; Good et al., 1993). Contextual constructivism and the related social constructivism hold that personal constructivism is too limited as humans are social beings, and knowledge creation is in uenced by the prior experiences and social environment of the student. Wheatley (1991, p. 49) summarises the position by claiming that ‘we continually negotiate the meaning of events in our lives so that we can beneŽ t from the experiences of others as well as our own’. Social constructivists believe that an important part of construction is social interaction through which we come to a common understanding of knowledge, including scientiŽ c concepts (Wheatley, 1991; von Glasersfeld, 1993; Solomon, 1994). Tobin and Tippins (1993) put it this way ‘the individual and social components [are seen as] being parts of a dialectical relationship where knowing is seen dualistically as both individual and social, never one alone, but always both’ (p. 20). However, we may appear to have the same view of concepts as others, but our understanding is commonly discrepant, for example, when there is an undetected communication breakdown (Johnson & Gott, 1996).

Data Collection The researchers began the larger study by conducting an in-depth analysis of the curriculum material that these students had encountered during their studies. This entailed a thorough examination of lesson plans, textbooks, lecture notes and other relevant documentation such as topic tests and Ž nal examinations. These data were synthesised into a summary of the mental models for chemical bonding. This summary comprised some 40 pages and a total of eight models along with extensions and modiŽ cations of the models. The models identiŽ ed, were, the sea of electrons model and band theory for metals, a model based on electron transfer and a further model involving the calculation of electrostatic charges for ionic substances, and the octet rule, the molecular orbital theory, the valence bond approach, and ligand Ž eld theory for covalent substances. Students’ views were elicited from semi-structured interviews according to the protocol detailed in Table I. This comprised a description of a given model chosen by the students during the interview, followed by probing of their understanding by the use of focus cards that depicted model use in some way (Table I). For example, for metals, this involved showing the participants samples of metallic substances (aluminum foil and steel wool), along with focus cards depicting the conductivity and malleability of a metal (examples of the cards for metallic bonding are provided in the Appendix). The student models revealed in interviews were then compared with the models found in the curriculum material; this revealed the alternative conceptions that form the focus of the present paper. Alternative conceptions were compiled into inventories and expert-validated by the teachers involved in this study. The expert’s principal contribution consisted of validation of the description of the models and of the researchers’ constructions and interpretation of the students’ alternative conceptions for chemical bonding reported here. It is

Metallic bonding Shown sample of aluminium foil—please describe the bonding in this substance Shown sample of steel wool—please describe the bonding in this substance Shown focus card depicting conductivity of copper wire—please explain this process Shown focus card depicting malleability of copper metal—please explain this process Shown focus card containing depictions of the bonding in lithium metal—which of these models appeals most/least to you? Ionic bonding Shown sample of sodium chloride—please describe the bonding in this substance Shown sample of lithium chloride—please describe the bonding in this substance Shown focus card depicting the conductivity of molten sodium chloride—please explain this process Shown focus card depicting the friability of sodium chloride—please explain this process Shown focus card containing depictions of the structure of sodium chloride—which of these models appeals most/least to you? Covalent bonding Shown sample of molecular iodine—please describe the bonding in this substance Shown sample of chloroform—please describe the bonding in this substance Shown focus card depicting reaction of copper chloride with ammonia solution—please explain this process Shown focus card containing a graphical depiction of the relationship between boiling point and Period number for a series of related hydrides—please explain this process Shown focus card containing depictions of the structure of benzene—which of these models appeals most/least to you?

TABLE I. Interview protocol for the study

Alternative Conceptions of Chemical Bonding Held by Students 175

176

R. K. Coll & N. Taylor

important to note that the research question in this work is subordinate to the aim of the larger study; namely, to establish the students’ preferred mental models for chemical bonding. Consequently, the results reported herein should not be deemed as an exhaustive catalogue of the students’ alternative conceptions for chemical bonding. Rather, these data represent a compilation of alternative conceptions revealed when the students discussed their mental models for chemical bonding. DeŽnition of Alternative Conceptions In this paper, a conception has been classiŽ ed as an alternative conception if it meets the following two criteria: the view was in disagreement with the scientiŽ c view and the conception was related to some aspect of chemical bonding, including speciŽ c details of a particular model for chemical bonding. For example, to describe the bonding in ionic compounds, the students typically described the size and shape of ions as well as the packing. Hence views at odds with the scientiŽ c view about ion size, shape or nature of packing, have been classiŽ ed as alternative conceptions However, during discussions centred on a focus card which compared the electrical conductivity of metallic copper with that of glass, there was evidence that some of the students held alternative conceptions about electrical conductivity. Because such alternative conceptions did not pertain to chemical bonding, they have not been included in the report of this work. Sample Description The study involved 30 participants from three academic levels; Year 12 secondary school students (age range 17–18 years), second and third year undergraduates (age range 19–20 years) and postgraduates (age range 23–26 years). The secondary school students formed two cohorts—two females (Natalie and Linda) from a private school in a middle class area of an Australian city, and four females (Anne, Anita, Claire and Frances) and four males (Neil, Keith, David and Richard) from single-sex schools in a middle class suburb of a small New Zealand city. The male secondary school students were, in general, less conŽ dent and outspoken than their female counterparts, although all students spoke freely during interviews. The participants were interested to pursue science-based careers and stated that they enjoyed chemistry. Eight of the undergraduate participants were intending BSc chemistry majors from a New Zealand university. Because the interviews for these undergraduates were conducted late in the year after the completion of lectures, undergraduate students had, at a minimum, completed 2 years of tertiary chemistry instruction. There were two male (Bob and Steve) and two female (Renee and Kim) second year undergraduate participants and two male (Alan and Mike) and two female (Jane and Mary) third year undergraduate participants. In addition to the New Zealand undergraduates there was one Australian male (Mike) and one female (Rosaline), both second year students from an Australian university. There were 10 postgraduate students, four New Zealand PhD candidates two being male (Jason and Kevin) and two female (Grace and Christine), and four MSc level candidates, two male (James and Brian) and two female (Jenny and Rose), from the same New Zealand university as the undergraduates. In addition there were two Australian PhD candidates, both male (John and Nigel). All the postgraduates were high academic achievers—a re ection of the entry requirements for postgraduate studies. In spite of this, there was a considerable spread in academic ability even within this cohort, with some students possessing outstanding academic records. All MSc candidates were purposefully chosen from the second year

Alternative Conceptions of Chemical Bonding Held by Students

177

class. The intention was to distinguish these postgraduates from Ž nal year BSc students: these selection criteria ensured that the masterate students had completed all of their MSc courses. Research Findings The alternative conceptions identiŽ ed in this study are detailed in Table II. Many of these alternative conceptions were not widely evident, several alternative conceptions being evidenced by a sole individual. Because the inquiry does not comprise a systematic attempt to uncover students’ alternative conceptions, a given alternative conception may have been identiŽ ed for only one student, but this does not necessarily mean only one student held this alternative conception. Strength and Weakness in Chemical Bonding The alternative conception that chemical bonding in metallic, ionic and covalent substances is weak was prevalent across all levels of the students (AC 1–3, Table II) who seemed unaware that this view con icted with direct physical evidence such as the hardness of metallic copper (Table II) and it is noteworthy that the students were shown a sample of thick solid copper metal at this point. Keith, for example, attributed the malleability of copper to weak bonding in the metal (AC 1), stating, ‘I think as it goes through rollers the copper atoms are just spread out. The bonding between them isn’t so strong’. David likewise reported, ‘I’d see the bonding as pretty loose, not strong bonding for copper’. Similar views were expressed about covalent bonding in some cases (AC 2), with David, for example, when discussing the bonding in chloroform (CHC13) stating, ‘it must be letting off some hydrogens and stuff, so I don’t think the bonding would be that strong’. The view that the electrostatic forces holding ionic compounds together are weak also was prevalent (AC 3, Table II). This view was the most common explanation offered by the students for the friability of ionic salts such as sodium chloride that was depicted in focus cards and was evident across all levels of student from secondary school to postgraduate. Neil stated ‘in the sodium chloride the force can just break it, the bonds are not very strong’, Steve said ‘the ionic bonding that is holding this crystal together here is weaker than is the case of the covalent’, and Brian ‘it’s much harder to break bonds, actual direct covalent bonds, than it is to break weak ionic bonds’. The other students, whilst not explicitly stating that such forces were weak, made statements that seemed to imply this. For example, Bob stated ‘the sodium chloride is only relying on an electrostatic force to hold together, whereas the other one’s got a lot of covalent bonds locking it rigidly together making it strong’. Christine apparently also believed ionic electrostatic forces in ionic substances such as sodium chloride are weak I see that bonding in sodium chloride can’t be as strong as in SiO2 although I see it; I guess I see the sodium chloride as different type of bonding. But if it’s going to crack it must have like little sections or something. That’s the only way I can explain it away, like if it’s the same type of bonding the silica one must be stronger otherwise the same thing would happen to it. Overall the data suggest that the students saw metallic and ionic bonding as weak or inferior to covalent bonding, although in a few instances covalent bonding was seen as weak also.

AC 1 Metallic bonding is weak bonding AC 2 Intramolecular covalent bonding is weak bonding AC 3 Ionic bonding is weak bonding AC 4 Continuous metallic or ionic lattices are molecular in nature AC 5 The bonding in metals and ionic compounds involves intermolecular bonding AC 6 The ionic radius of the sodium ion is greater than the chloride ion AC 7 The ionic radius of the lithium ion is greater than the sodium ion AC 8 Polar covalent compounds contain charged species AC 9 Molecular iodine contains 1 minus ions AC 10 The charged species in metallic lattices are nuclei rather than ions

TABLE II. Students’ alternative conceptions (AC) identiŽ ed in the study AC 11 Metallic lattices contain neutral atoms AC 12 Electronegativity comprises attraction for a single electron AC 13 Molecular iodine is metallic in nature AC 14 Ionic bonding comprises sharing of electrons AC 15 Ionic and metallic bonding contain an element of directionality AC 16 Ions in close-packed metal lattices possess other than eight nearest neighbours AC 17 Metal to non-metal bonding in alloys is electrostatic in nature AC 18 Ionic shape and packing is inuenced by pressure AC 19 Intermolecular forces are inuenced by gravity AC 20 Glass is an ionic crystalline substance

178 R. K. Coll & N. Taylor

Alternative Conceptions of Chemical Bonding Held by Students

179

Molecularity in Continuous Ionic and Metallic Structures Previous work revealed that the undergraduate and postgraduate students seemed to possess a greater appreciation of the continuous nature of ionic and metallic lattices than secondary school students (Coll & Treagust, in press). Interestingly, some of the students across all levels possessed alternative conceptions about the nature of metallic and ionic lattices, seeming to believe that such lattices were molecular in nature or contained distinct molecular species (AC 4, Table II). For example, the students used the term molecules to describe particles in metals and ionic substances. Frances, attempted to explain the lack of conductivity of molten sodium chloride thus, ‘the molecules can only move a little bit side to side so they can’t move and make the bulb glow’ and Keith stated ‘the bonds are rigid and aren’t allowing the sodium chloride to ionise and this holds the electrons from  owing on through the sodium because the molecules are held tightly together by the bonds in the sodium chloride, so they can’t move around’. Keith also introduced the concept of molecules when describing the conductivity of metallic copper stating, ‘the copper molecule, it’s  owing. Like it’s going from positive to negative so the electrons can  ow along’ and Bob when describing a ball-and-stick depiction of bonding in the metal lithium stating ‘I like the way you can see the packing together of the molecules in a regular fashion’. Some caution is necessary in the interpretation of these data as it is possible that these statements merely represent inappropriate or careless use of nomenclature. However, some of the students identiŽ ed speciŽ c molecular species in metals and ionic substances, suggesting that their use of the term molecule is purposeful, as for example, in the description of the structure of sodium chloride and caesium chloride by Alan, ‘the caesium atom makes contact with all these chlorides in the middle. That sort of interaction between the caesium and chloride, like how you are taught NaCl, the NaCl molecule’. A related alternative conception, that metals and ionic compounds possess intermolecular bonds, was evident for three students (AC 5, Table II). David stated ‘I suppose there is always the van der Waals things with just the attraction of the electrostatics and stuff’, and as seen in a drawing (Fig. 1) and writing used by Keith to describe the bonding in the ionic substance LiCl. Then there is bonding between another, and another [draws wavy lines between units of LiCl, writes van der Waals next to wavy line]. The notion of molecularity may have led to this alternative conception, in that the students felt it necessary to explain why the structure was held together, and drew upon the concept of van der Waals forces to do so. Alternative Conceptions of Ion Size There were two remarkably prevalent alternative conceptions related to the size of ions identiŽ ed; one in which the sodium ion was viewed as being larger in size than the chloride (AC 6, Table II), the other that the lithium ion was larger than the sodium ion (AC 7, Table II). Again these views were held across all three academic levels. Many of the students stated that sodium was the larger ion of the two in sodium chloride. Anita stated ‘the sodium would be bigger than chlorine’, Neil ‘lithium’s smaller than sodium and, and chloride is smaller than sodium’ and Richard ‘I would say the sodium is the large one and then the smaller ones would be the chloride’. Likewise, Neil said ‘lithium would be bigger it’d have more protons and electrons it’d be bigger’. Alan stated ‘I would

180

R. K. Coll & N. Taylor

FIG. 1. Keith’s drawing illustrating the bonding in lithium chloride (LiCl).

consider lithium to slightly larger than the sodium’, and Mary said ‘the lithium maybe a little bit bigger than sodium’. It is noteworthy that the respondents were in possession of a Periodic Table at the time. The responses above reveal the tentative nature of the students’ views when the size difference is described at being ‘slightly larger’ and ‘maybe a little bit bigger’. Views expressed by some of the secondary school students offered clues to the origin of the alternative conception that the chloride ion is smaller than the sodium ion. It seems that students at this level at least, may have confused ionic size with the Periodic trend in atomic size. David: I would say the Na would be bigger and the Cl would be smaller, I imagine they would be. Interviewer: Why do you see it that way? David: The explanation I have had of that is that as you move across the Periodic Table, well there’s more electrons in the same shell with an increasing number of protons as well, and that like attracts them closer which just means it is smaller. It is common to illustrate the structure of ionic substances such as sodium chloride in lectures and curriculum material using space-Ž lling and ball-and-stick models; informal interviews of all of the instructors in this inquiry indicated that this formed part of their teaching strategy; likewise, the textbooks used by the students contained diagrams that showed the ions to be similar in size. The ball-and-stick models often have the ion size depicted as similar; whereas there is a signiŽ cant difference in ion size in space-Ž lling models. Alternative Conceptions for the Ionic and Covalent Character of Chemical Bonds As mentioned above the inappropriate use of nomenclature may be instrumental in the formation of some alternative conceptions, and this is particularly evident in the case of the rather esoteric notion of the ionic–covalent continuum. Jason, one of the most academically able and most erudite of the postgraduate participants in the study, exempliŽ es the situation. His academic ability coupled with his extensive tutoring of undergraduate chemistry suggest that he is likely to be aware of the importance of nomenclature. Upon describing the formation of the chloride ion from neutral chlorine, he clearly showed that he understands the difference between an ion and a neutral atom,

Alternative Conceptions of Chemical Bonding Held by Students

181

that is, chlorine and chloride. However, he frequently made inappropriate use of nomenclature in subsequent explanations. I see the chlorine as being not a chlorine atom but a chlorine ion, so a chlorine which has gained an electron from the sodium atom so that the chlorine atom has a negative charge. The sodium has a positive charge. It’s like this structure is essentially held together by electrostatic interactions. Now there’s also repulsions as well because you also have another chlorine atom which is reasonably close to that chlorine atom so there’s repulsions between the two. Because of his expertise and experience, it seems that Jason is clear in his own mind of the difference between the terms chlorine and chloride; thus interchanging the terms may not be particularly detrimental for his understanding of chemical bonding. However, it is possible that the interchanging of such terms for novices may lead to alternative conceptions. To illustrate, consider another highly prevalent alternative conception found—that polar covalent compounds contain charged species (AC 8, Table II). Frances indicated that chloroform contains a proton, and Anita identiŽ ed Cl minus in the same compound. Frances: Because these two have Hs in them and H with the Cl, and the highly electronegative ones like chlorides and things like that that gives the hydrogen bonding whereas they can readily donate a proton away to the hydrogen. Anita: Yeah because H is plus and Cl is minus, but I don’t know why, I just know it is different but I don’t know why. Similarly, Mary, an undergraduate, stated explicitly that she viewed the bonding in hydrogen halides as ionic in nature, rather than polar covalent, or possessing of some ionic character, stating ‘I think of them as being ionic’. Jenny, a postgraduate, articulated her views about the bonding in chloroform in greater detail drawing on previous experience of the bonding in carboxylic acids; part of her MSc research project. However, the value of doing this seems dubious, as she confused the donation of protons in carboxylic acids with a perception of lability of the chlorine atom in chloroform (CHCl3). The fact that she used the term chloride, rather than chlorine, may indicate that inappropriate nomenclature has in part contributed to her alternative conception. I know that to remove an atom from chloroform is quite difŽ cult. But if you look at other organic molecules will, um … I mean if you think, like your carboxylic acid it’s quite easy to remove the proton and the same for the chloride, like a halide. It’s quite easy to remove chlorine from a long sort of carbon chain. The fact that Jenny described the chlorine in chloroform as chloride, indicative of an ionic species, may have led her to make an inappropriate link to her own experience with the ionisation of carboxylic acids. An examination of interview transcripts revealed that it was routine across all academic levels for the students to interchange terms pertaining to halogens with the charged halide. Students used the terminology for neutral atoms when describing the bonding in ionic compounds, and similarly described halides as halogens, as seen in Keith’s description of the bonding in sodium chloride and Steve’s for lithium chloride. Keith: OK the sodium has got one electron in its outer shell, and the chlorine has got seven. So the chlorine requires one more electron. The bonds in the sodium chloride I don’t think are as strong between the sodium and the

182

R. K. Coll & N. Taylor chlorine so they are not attracted to each other, the sodium and the chloride ions.

Steve: Well basically it’s going to be easier for the chlorine’s wait a sec., um because the chlorine’s are still, the chloride’s have still got the same size. An underlying origin of this view may be misunderstandings about the notion of the ionic–covalent continuum. The students are told that there is no such thing as pure ionic bonding and that atoms in polar covalent compounds hold partial charges. It is possible that this notion reinforces the view in the students’ minds that chlorine is charged, causing confusion between the notion of a partially charged chlorine atom in a polar covalent compound and the negative chloride ion. In support of this proposition, inappropriate use of terminology seems to be in uential in the formation of alternative conceptions about the bonding in pure molecular covalent substances like molecular iodine (I2) (AC 9, Table II). It is perhaps more likely that the students confuse chloride and chlorine in a substance like chloroform which they are told is polar covalent and where the chlorine atom carries a partial charge. However, this is not the case for non-polar molecular covalent species like homonuclear diatomics such as molecular iodine. The fact that the same alternative conception is seen for such substances adds credence to this view. Mary and Rose seemed to believe that the iodine atoms in I2 carry a negative charge, although they both subsequently stated that the bonding comprised sharing of electrons, that is, a covalent bond. Mary: Iodine well I know iodine as being I2 that’s how I remember it. Iodine is, um, I minus so therefore it’s got one extra electron I guess sitting there, that’s not doing anything [respondent laughs] and it’s able to pair up with another iodine which also has a spare, and those two electrons come together and form a bond. Rose: In iodine, well it’s like two atoms of iodine, they would be equally contributing from the covalent bonds because each I minus, each iodine would be like lacking one electron. So they join together to form a more stable I2, donating an electron each and they are shared between the two atoms. Interviewer: So that’s a combination of two I minuses is it? Rose: Yeah. In a similar way students seemed to confuse ions with neutral species and nuclei in metallic bonding (AC 10–11, Table II). Two alternative conceptions of this nature were identiŽ ed, namely, that metallic lattices contain neutral atoms and that the positive ions present are nuclei. Interviewer: Do you see this here as being lithium plus here? Steve: No I don’t. I just see that as just being a lithium. Mary: I guess that’s the nucleus [indicating the 1 symbols in the sea of electrons model, see focus card MB01 in the Appendix]. Kevin: Well my conception anyway is the fact that the charges of all the nuclei are very positively charged. The electrons are negatively charged. That will make the whole thing sort of stay together. Another possible origin of this alternative conception maybe the nature of visual clues used in the diagrams depicting the sea of electrons model, in which spheres enclose a positive sign (focus card MB01, Appendix).

Alternative Conceptions of Chemical Bonding Held by Students

183

Alternative Conceptions about Electronegativity It is possible that another alternative conception, that electronegativity comprises attraction for a sole electron (AC 12, Table II) may also be related to this. Steve: It’s the electron attracting ability of a species. So in the case of  uorine for instance being the most electronegative element, because it has only one space left to Ž ll in its 2p orbital, then it would very strongly attract an electron to Ž ll that orbital, to attain a stable Ž lled valance orbital. Claire: It’s still ionic because the lithium has a stronger attraction. Interviewer: Stronger attraction to what? Claire: To the lithium. So the one electron they are sharing it will still spend most of the time around the chlorine or chloride. Although a deŽ nition of electronegativity was not explicitly elicited, if a given student introduced the term, he or she was asked to explain what the term meant to them. Electronegativity is not an aspect of chemical bonding per se. However, students’ understanding of electronegativity impacts upon their understanding of bonding. Most of the students produced a deŽ nition of electronegativity that was in general agreement with the scientiŽ c view. However, Steve seemed to believe that the attraction was for a single electron rather than a greater attraction for the shared pair. His statement that there is ‘only one space left’ resulting in greater attraction suggests that the Ž lled valence shell concept of the octet rule is in uential in his view of electronegativity. Alternative Conceptions about Metal and Non-metal Nature An interesting alternative conception revealed in the study was that molecular iodine (I2) is metallic in nature (AC 13, Table II). Christine and Alan stated that they viewed iodine as metallic and Alan also related the bonding to that of lithium which he previously encountered in a focus card, Christine stating ‘I deŽ nitely think of it as a metal’ and Alan ‘perhaps even more like the structure of that lithium. I sort of see lithium–lithium bonds. It’s the same in the case of the iodine’. Jason clearly considered that I2 contains covalent bonds: Iodine is kind of metallic, so I guess iodine certainly is a gas. Iodine will sublime when you heat it up and you will see the purple colouring in the air and that’s iodine gas and iodine is I2. So you have one iodine covalently bonded to another iodine [drawing two Is inside circles linked together, Fig. 2] so you have I—I like that. So my guess it’s kind of metallic. Jason stated that ‘iodine has some sort of metallic properties’. The origins of this view may lie in the lustrous, rather metallic, appearance of the crystalline appearance of the sample used during the interviews. Upon probing it seemed that Jason did not believe molecular iodine is a metal, yet he persisted with his statement that it will possess metallic properties. Interviewer: What makes you say it has metallic qualities? Jason: Just looking at it. It looks metallic it’s somewhat shiny. It’s even grey in

184

R. K. Coll & N. Taylor

FIG. 2. Jason’s drawing illustrating the bonding in molecular iodine (I2).

colour and you have sort of discrete little crystals of it. I don’t know. When I say it looks like a metal, I don’t expect it to behave as a metal. Interviewer: So you are saying it is metallic in appearance? Jason: Yeah. It obviously has some sort of metallic characteristics to it. Jason seems oblivious to the contradiction in his account. On one hand he states ‘I don’t expect it to behave as a metal’ and then ‘it obviously has some sort of metallic characteristics’. His view may be in uenced by notions of Periodic trends. Although this has not been articulated by Jason, it is common to state that there is an increase in metallic character as one goes down a group in the Periodic Table (Chang, 1991, p. 167). This, along with the metallic appearance of I2 may be the cause of the alternative conception. An interesting alternative conception was related to the bonding in ionic compounds like sodium chloride (AC 14, Table II). A number of the students seemed to become confused between ionic bonding and covalent bonding. Two secondary school students, Anne and Frances, identiŽ ed the bonding in ionic compounds as covalent, despite initially describing a process of electron transfer. Anne stated ‘the electrons are sharing between the caesium and the chloride’ and Frances likewise stated ‘the chloride is electronegative which means it can give away an electron and Na plus means it can accept the electron’. The comments made by Frances in particular provide clues to the origins of this alternative conception. Frances mentioned electron transfer, but went on to state that the bonding is covalent in nature. Such a description possesses elements of the theory employed to explain the ionic–covalent continuum. Other Alternative Conceptions about Chemical Bonding In addition to the more prevalent alternative conceptions described above, there were a number of conceptions that were identiŽ ed for a sole participant. Jason seemed confused about the number of nearest neighbours in copper and stated that the malleability of metallic copper involved changes to the number of nearest neighbours; he saw 12, then six, followed by four neighbours, instead of the eight that are actually present in copper metal. Steve seemed confused about the nature of bonding in the alloy steel (AC 17, Table II), viewed the bonding between the non-metal carbon and iron as electrostatic in nature rather than covalent. Well you have still got the retention of the electrostatic forces between the iron cation and the electrons. But, I am just trying to think back. With the introduction of the carbon, there’s actually going to be new centres located inside the metal structure. I presume that they would be held by some sort of electrostatic interaction. Rose, a postgraduate, seemed to believe that the shape of ions could be in uenced by macroscopic factors such as pressure (AC 18, Table II) ‘like it’s maybe the conditions it’s

Alternative Conceptions of Chemical Bonding Held by Students

185

under. The pressure might make a difference’. Neil possessed a novel view of intermolecular forces, believing that they were affected by gravity (AC 19, Table II) ‘it moves a little bit, but it must be that the charges on each side would not be as strong I would imagine. It may be also because it’s affected more by gravity that the others’. In addition, Steve seemed to believe glass was a crystalline substance stating ‘glass is basically a silica structure, that’s silicon covalently bonded in a silica structure with SiO4 units’. Although these views were only expressed by one individual, as mentioned previously, it is possible that these views are not idiosyncratic and were also held by the other students. Summary The interview data revealed prevalent alternative conceptions for chemical bonding across three levels of learning. In addition to some common alternative conceptions previously reported in the literature, such as misunderstandings about intermolecular forces and molecularity of continuous lattices, the inquiry found a surprising number of alternative conceptions about simple ideas like ion size and shape. Some 20 alternative conceptions were revealed, the most common being belief that continuous ionic or metallic lattices were molecular in nature, and confusion over ionic size and charge. Implications for Teaching and Learning It is surprising that the undergraduate and postgraduate students, especially given their good academic records, held such alternative conceptions and it is interesting to consider why, in particular, the undergraduate and postgraduate students became confused about a number of fairly straightforward concepts. In one statement John, one of the PhD students, described his views about the purpose of tertiary education. He made it plain that he had a great desire to gain what he perceived as highly practical skills from his university studies. He stated that he chose his particular tertiary institution deliberately because he felt it provided ‘a more hands-on approach, something that is a lot more practical’. He reinforced this view later when questioned about the concept of antibonding which he had introduced during a discussion regarding the bonding in benzene. All I remember is that they are silly things that stick out either side. I couldn’t really understand or fathom them. That’s just how it is. That’s how I remember the antibonding things. I mean just drawing your diagrams you’ve just got to remember your antibonding. You have your bonding and you have antibonding orbitals. Antibonding orbitals are just there because they are there. That’s all it is. I didn’t really understand why they are there, just there so the numbers balance. It seems that, John at least, is interest in theoretical aspects of bonding models when they have a practical application. The fact that he dismisses antibonding orbitals as ‘silly things’ suggests it is possible that students ascribe complex conceptions low status unless they can see their relevance to their work and subsequent study. However, the origins of alternative conceptions are many fold. Of prime importance are students’ existing views about abstract concepts. It is common for tertiary teachers to consider that students have little or no knowledge of such concepts, other than rudimentary grounding in simple theories such as the sea of electrons model or the octet rule. Certainly at the tertiary institutions involved in this study, the teachers assume no foreknowledge of, for example, molecular orbital theory or the valence bond approach. However, although

186

R. K. Coll & N. Taylor

students may not have encountered such theories explicitly, they may harbour other alternative conceptions that impact upon their ability to understand such complex and abstract models. The students in this study seemed to Ž nd great difŽ culty remembering details of models and concepts despite, in some cases, having encountered them in relatively recent instruction. The amount of material that the students encounter in an undergraduate science course is formidable. This, coupled with the mass of advanced material encountered during their postgraduate studies, may offer some explanation for the surprising number of alternative conceptions revealed for the advanced level students in this study. It would seem reasonable to conclude that confusion and careless use of terminology is in uenced by the sheer mass of material the students encounter at the tertiary level. We have argued previously that it is not feasible to consider removing the teaching of these models from the undergraduate chemistry curriculum (Coll & Treagust, in press). Modern chemistry teachers at the senior secondary and tertiary level face a tension in the competing aims of teaching, that is, the desire to provide adequate chemistry knowledge for those seeking specialist careers as chemists, typically chemistry majors, and yet avoiding overloading non-specialists with unnecessary material that will be of limited value in their own studies and subsequent careers (Fensham, 1980). However, tertiary chemistry teachers may wish to consider the advisability of limiting the teaching of such models until an advanced stage of the undergraduate degree, say the third year of a conventional 3 year bachelors degree. This proposition is offered as chemistry non-majors will have little need for models in their subsequent studies. Hence, the value of teaching, for example, biology majors, sophisticated complex and highly abstract mental models is in our view dubious.

Correspondence: Dr Richard K. Coll, School of Science and Technology. The University of Waikato, Private Bag 3105, Hamilton, New Zealand. E-mail: [email protected]. REFERENCES ABRAHAM, M.R., GRZYBOWSKI, E.B., RENNER, J.W. & MAREK, E.A. (1992) Understandings and misunderstandings of eighth graders of Ž ve chemistry concepts found in chemistry textbooks, Journal of Research in Science Teaching, 29, pp. 105–120. BARROW, G.M. (1994) General chemistry and the basis for change, Journal of Chemical Education, 71, pp. 874–878. BIRK, J.P. & KURTZ, M.J. (1999) Effect of experience on retention and elimination of misconceptions about molecular structure and bonding, Journal of Chemical Education, 76, pp. 124–128. BOO, H.K. (1998) Students’ understandings of chemical bonds and the energetics of chemical reactions, Journal of Research in Science Teaching, 35, pp. 569–581. BRADLEY, J.D. & BRAND, M. (1985) Stamping out misconceptions, Journal of Chemical Education, 62, p. 318. BUTTS , B. & SMITH, R. (1987) HSC chemistry students’ understanding of the structure and properties of molecular and ionic compounds, Research in Science Education, 17, pp. 192–201. CARTER , C.S. & BRICKHOUSE, N.W. (1987) What makes chemistry difŽ cult? Alternate perceptions, Journal of Chemical Education, 66, pp. 223–225. CHANG, R. (1991) Chemistry, 4th edn (New York, McGraw-Hill). COBERN, W.W. (1993) Contextual constructivism: the impact of culture on the learning and teaching of science, in: K. TOBIN (Ed.) The Practice of Constructivism in Science Education (Hillsdale, NJ, Lawrence Erlbaum Associates). COLL, R.K. & TREAGUST , D.F. (in press) Students’ preferred mental models for chemical bonding, Research in Science Education.

Alternative Conceptions of Chemical Bonding Held by Students

187

DE POSADA, J.M. (1997) Conceptions of high school students concerning the internal structure of metals and their electric conduction: structure and evolution, Science Education, 81, pp. 445–467. DRIVER, R. (1981) Pupil’s alternative frameworks in science, European Journal of Science Education, 3, pp. 93–101. DRIVER, R. (1989) Students’ conceptions and the learning of science, International Journal of Science Education, 11, pp. 481–490. DRIVER, R. & EASLEY, J. (1978) Pupils and paradigms: a review of literature related to concept development in adolescent science students, Studies in Science Education, 5, pp. 61–84. FENSHAM, P.J. (1980) Constraint and autonomy in Australian secondary science education, Journal of Curriculum Studies, 12, pp. 186–206. FENSHAM, P.J. & KASS, H. (1988) Inconsistent or discrepant events in science instruction. Studies in Science Education, 15, pp. 1–16. GABEL, D. (1998) The complexity of chemistry and implications for teaching, in: B.J. FRASER & K.G. TOBIN (Eds) International Handbook of Science Education (Dordrecht, Kluwer). GARNETT, P.J. & T REAGUST, D.F. (1992) Conceptual difŽ culties experienced by senior high school students of electrochemistry: electric circuits and oxidation–reduction equations, Journal of Research in Science Teaching, 29, pp. 121–142. GILBERT, J.K. & WATTS, D.M. (1983) Concepts, misconceptions and alternative conceptions: changing perspectives in science education, Studies in Science Education, 10, pp. 61–98. GOH, N.K., KHOO, L.E. & CHIA, L.S. (1993) Some misconceptions in chemistry: a cross-cultural comparison, and implications for teaching, Australian Science Teachers Journal, 39, pp. 65–68. GOOD, R.G., WANDERSEE, J.H. & ST JULIEN , J. (1993) Cautionary notes on the appeal of the new “ism” (constructivism) in science education, in: K. TOBIN (Ed.) The Practice of Constructivism in Science Education (Hillsdale, NJ, Lawrence Erlbaum Associates). GRIFFITHS, A.K. & PRESTON, K.R. (1989) An investigation of grade 12 students’ misconceptions relating to fundamental characteristics of molecules and atoms. Paper presented at the 62nd conference of the National Association for Research in Science Teaching, San Francisco. HARRISON, A.G. & TREAGUST , D.F. (1996) Secondary students’ mental models of atoms and molecules: implications for teaching chemistry, Science Education, 80, pp. 509–534. HAWKES , S.J. (1996) Salts are mostly NOT ionized, Journal of Chemical Education, 73, pp. 421–423. JOHNSON, P. & GOTT, R. (1996) Constructivism and evidence from children’s ideas, Science Education, 80, pp. 561–577. KIRKWOOD, V. & SYMINGTON , D. (1996) Lecturer perceptions of student difŽ culties in a Ž rst-year chemistry course, Journal of Chemical Education, 73, pp. 339–343. NAKHLEH, M.B. (1992) Why some students don’t learn chemistry: chemical misconceptions, Journal of Chemical Education, 69, pp. 191–196. PETERSON, R.F. & T REAGUST, D.F. (1989) Grade-12 students’ misconceptions of covalent bonding and structure, Journal of Chemical Education, 66, pp. 459–460. PETERSON, R.F., TREAGUST , D.F. & GARNETT, P. (1989) Development and application of a diagnostic instrument to evaluate grade-11 and grade-12 students’ concepts of covalent bonding and structure following a course of instruction, Journal of Research in Science Teaching, 26, pp. 301–314. PFUNDT, H. & DUIT, R. (1997) Bibliography: student’s alternative frameworks and science education, 4th edn (Kiel, University of Kiel). SANGER, M.J. & GREENBOWE, T.J. (1997) Common student misconceptions in electrochemistry: galvanic, electrolytic, and concentration cells, Journal of Research in Science Teaching, 34, pp. 377–398. SCHMIDT, H.-J. (1997) Students’ misconceptions—looking for a pattern, Science Education, 81, pp. 123–135. SOLOMON, J. (1994) The rise and fall of constructivism, Studies in Science Education, 23, pp. 1–19. STAVY, R. (1995) Conceptual development of basic ideas in chemistry, in: S.M. GLYNN & R. DUIT (Eds) Learning Science in the Schools: research reforming practice (Mahwah, NJ, Lawrence Erlbaum Associates). TABER, K.S. (1994) Misunderstanding the ionic bond, Education in Chemistry, 31, pp. 100–102. TABER, K.S. (1995a) An analogy for discussing progression in learning chemistry, School Science Review, 76(276), pp. 91–95. TABER, K.S. (1995b) Development of student understanding: a case study of stability and lability in cognitive structure, Research in Science and Technological Education, 13, pp. 89–99.

188

R. K. Coll & N. Taylor

TABER, K.S. (1997) Student understanding of ionic bonding: molecular versus electrostatic framework, School Science Review, 78(285), pp. 85–95. TABER, K.S. (1998) An alternative conceptual framework from chemistry education, International Journal of Science Education, 20, pp. 597–608. TAYLOR, N. & COLL, R.K. (1997) The use of analogy in the teaching of solubility to pre-service primary teachers, Australian Science Teachers’ Journal, 43, pp. 58–64. TAYLOR, N. & LUCAS, K. (1997) The trial of an innovative science programme for preservice primary teachers in Fiji, Asia-PaciŽc Journal of Teacher Education, 25, pp. 325–343. TOBIN, K., BRISCOE, C. & HOLMAN, J.R. (1990) Overcoming constraints to effective elementary science teaching, Science Education, 74, pp. 409–420. TOBIN, K. & TIPPINS, D. (1993) Constructivism: a paradigm for the practice of science education, in: K. TOBIN (Ed.) The Practice of Constructivism in Science Education (Hillsdale, NJ, Lawrence Erlbaum Associates). VON GLASERSFELD, E. (1993) Questions and answers about radical constructivism, in: K. TOBIN (Ed.) The Practice of Constructivism in Science Education (Hillsdale, NJ, Lawrence Erlbaum Associates). WHEATLEY , G.H. (1991) Constructivist perspectives on science and mathematics learning, Science Education, 75, pp. 9–21. ZOLLER, U. (1990) Students’ misunderstandings and misconceptions in college freshman chemistry (general and organic), Journal of Research in Science Teaching, 27, pp. 1053–1065.

Alternative Conceptions of Chemical Bonding Held by Students Appendix

189

190

R. K. Coll & N. Taylor

Alternative Conceptions of Chemical Bonding Held by Students

191