Electrostatic work brings opposite charge carriers together, while nonelectrostatic work causes charge separation. To maintain this important distinction, we shall avoid the term voltage.
First of all, using the language of "causation" in this context is incorrect, as discussed in item #79.
Secondly, the idea that "charge separation" is relevant violates a fundamental principle of physics: Force is not motion. The electron in an atom spends half its time moving toward regions of more-positive voltage, and half toward more-negative voltage. Surely you don't mean to suggest that one direction is EMF and the other is potential difference.
Thirdly, the distinction between electrostatic and nonelectrostaic voltages is not "important" ... and indeed is not even real. Consider a black box containing a capacitor (with voltage V) plus a battery (with voltage V) and a dynamo (putting out voltage V) all hooked in parallel. You apply a voltmeter to the terminals of the black box, without looking inside. Tell me, what are you measuring? EMF?? Potential difference?? Both?? Neither??
I claim the physics does not recognize the distinction. The voltmeter measures voltage, which is energy per unit charge, period. In this context as in so many others, as discussed in item #35, energy is energy, period. The more ways of dividing it up, the more ways of making a mistake.
The notion that the battery voltage is nonelectrostatic is not viable, as discussed in item #62. This is further evidence that dividing the energy into EMF and potential-difference is not "important". If you can divide it incorrectly without noticing, it's not "important".
The idea that a battery is the prototypical potential difference while a battery is the prototypical EMF is very strange. Every circuit designer I've ever met models a battery as a high-value capacitor with a somewhat squirrelly transient response.
Taking a look at the larger picture: We know what the right answer looks like. We know the Maxwell equations. We know about electric fields. We can integrate the electric field along some path to get the voltage. We know about Thévenin equivalent open-circuit voltage. This is all completely standard and known to work.
However, an elaborate alternative structure of terminology and concepts has been set up. Usually when somebody goes to that much trouble, there's a reason ... but in this case I cannot imagine what the reason could be. The structure doesn't fit the facts. In the book, the definition of potential difference doesn't fit the way the term is used, as discussed in item #60. In the book, the definition of EMF doesn't fit the way the term is used, as discussed in item #62. The arguments in favor of EMF don't fit the facts. The arguments against voltage don't fit the facts. I'm mystified. Why go to all that trouble?
In section 26.6 on page 702 the footnote says
First of all, using the language of "causation" in this context is incorrect, as discussed in item #79.
Secondly, the idea that "charge separation" is relevant violates a fundamental principle of physics: Force is not motion. The electron in an atom spends half its time moving toward regions of more-positive voltage, and half toward more-negative voltage. Surely you don't mean to suggest that one direction is EMF and the other is potential difference.
Thirdly, the distinction between electrostatic and nonelectrostaic voltages is not "important" ... and indeed is not even real. Consider a black box containing a capacitor (with voltage V) plus a battery (with voltage V) and a dynamo (putting out voltage V) all hooked in parallel. You apply a voltmeter to the terminals of the black box, without looking inside. Tell me, what are you measuring? EMF?? Potential difference?? Both?? Neither??
I claim the physics does not recognize the distinction. The voltmeter measures voltage, which is energy per unit charge, period. In this context as in so many others, as discussed in item #35, energy is energy, period. The more ways of dividing it up, the more ways of making a mistake.
The notion that the battery voltage is nonelectrostatic is not viable, as discussed in item #62. This is further evidence that dividing the energy into EMF and potential-difference is not "important". If you can divide it incorrectly without noticing, it's not "important".
The idea that a battery is the prototypical potential difference while a battery is the prototypical EMF is very strange. Every circuit designer I've ever met models a battery as a high-value capacitor with a somewhat squirrelly transient response.
Taking a look at the larger picture: We know what the right answer looks like. We know the Maxwell equations. We know about electric fields. We can integrate the electric field along some path to get the voltage. We know about Thévenin equivalent open-circuit voltage. This is all completely standard and known to work.
However, an elaborate alternative structure of terminology and concepts has been set up. Usually when somebody goes to that much trouble, there's a reason ... but in this case I cannot imagine what the reason could be. The structure doesn't fit the facts. In the book, the definition of potential difference doesn't fit the way the term is used, as discussed in item #60. In the book, the definition of EMF doesn't fit the way the term is used, as discussed in item #62. The arguments in favor of EMF don't fit the facts. The arguments against voltage don't fit the facts. I'm mystified. Why go to all that trouble?
What's the problem we're trying to solve here?