Heat–Cool: A Simpler Differential Scanning Calorimetry Approach for Measuring the Specific Heat Capacity of Liquid Materials

Round 1

Reviewer 1 Report

The manuscript presents a novel approach to obtain liquid heat capacity data using a differential scanning calorimeter.  My questions and concerns follow:

1. Although the authors give some explanation, it is still unclear why a calibration constant/corrective factor is not needed with this method.  This could benefit from additional explanation.

2. Equation 6 - Using "delta" as a variable is confusing, since it is used in connection with other variables.  Recommend that another variable be used.

3. Heating rates are easier to certify than cooling rates.  Uncertainty will likely be different for these two quantities.

4. The discussion (in Experimental and Conclusion) about the need to have large samples and eliminate headspace is troubling.  This is contrary to our DSC experience.  At the conditions of the reported measurements, the vaporization effects should be small.

5. The data in the manuscript are for a very limited number of compounds and over a limited temperature range.  This diminishes the strength of the claims about the method.  I recommend more compounds with a variety of functional groups and a broader range of temperatures.  It would also be valuable to compare to certified temperature-dependent correlations from NIST and DIPPR.  Also, the purity of the chemicals should be stated.

6. The incorrect temperature dependence for water is troubling, and this too diminishes the claims in the manuscript.

In conclusion, the method is intriguing and seems to show promise. However, there are concerns with some aspects of the methodology and the limited data do an inadequate job of certifying the method.

 

Author Response

Specific responses to reviewer 1 listed by reviewer’s paragraph number.

1. The justification for the absence of using a corrective factor to improve results lies in what is desired for the bottom line results. If agreement with literature in the 1 -2 % range is adequate for the purpose of the study, then adding a correction/calibration factor is unnecessary and merely adds additional experimental time. If the suggested the 1-2 % range is not sufficient, then a correction factor can be included. This study demonstrates agreement in the 1-2 % range.
2. The symbol D has been replaced by s throughout the mss.
3. As to different uncertainties for heating and cooling rates, the differences implied by the specifications of the manufacturer are negligible, certainly over the modest temperature ranges of the experiment. This is one of excellent features of the TA Tzero baseline control.  It could be that this method would be impossible on a DSC instrument of different design.  The authors have not tested the method on their PE Model 7 DSC.
4. A large sample size does reduce problems with vaporization and measuring the c_p of two phases. Preliminary experiments showed that smaller samples in the Tzero sample pans would often give inconsistent results. Moving to larger samples that came close to filling the pans provided much more reproducible results.  Also as the heat capacities of liquids are generally small, certainly compared to solids, the larger sample provided greater detectable signals. The purity of the chemicals used is now provided in the Materials table.
5. While the recommendation of studying more compounds over different temperature ranges is certainly a good one, for the authors, this is not possible. The undergraduate coauthors have now graduated, pursing jobs and graduate degrees. I am retired.  These facts mean further experiments would fall to other workers who see merit in the method and would seek to extend and even improve it.
6. The apparent incorrect temperature dependence of water c_p might be understood by looking at the experimental values and their estimated uncertainties. As now mentioned in the text, the instrument might not be sensitive enough to detect the really small changes in water’s c_p over the temperature range studied.

Reviewer 2 Report

The manuscript describes the use of DSC in a heating/cooling loop, which can simplify heat capacity measurements by removing  the need for calibration experiments. This is potentially very useful.

I have one concern. Figure 1 shows the principle quite clearly but is not entirely accurate, in that, while the heat flows are measured over the same temperature range, they are not run simultaneously. The point is that the baseline for the cooling loop is the extension of that for the for the heating is loop. As long as the baseline is constant, as in Figure 1, there is no problem but if the baseline is curved or has a slope, then the situation changes.

I think that the authors could profitably consider this in the manuscript.

It would also be useful to the reader for the authors to explain the multiplication of Delta by 3. To me at least this isn't immediately obvious.

One small typo, in the opening sentence of the Introduction should read: ...assumed that the differences in the physical properties of the sample and ....

Author Response

Responses to reviewer 2 that follow the order the paragraphs.

1. As to different uncertainties for heating and cooling rates, that is baselines, the differences implied by the specifications of the manufacture are negligible, certainly over the modest temperature ranges of the experiment. This is one of excellent features of the TA Tzero baseline control.
2. The explicit explanation for the factor of 3 in the c_p calculation is now included in the sample calculation.
3. The poorly chosen preposition has been replaced by the reviewer’s suggestion.

Reviewer 3 Report

The manuscript entitled “ Heat-Cool: A novel and simpler differential scanning calorimetry approach for measuring the specific heat capacity of liquid materials including as functions of temperature over small temperature ranges” addresses a very serious and important issue in calorimetry. The determination of an absolute and reliable value of the specific (or molar) heat capacity over a wide temperature range is the “high school” in calorimetry. Therefore, any new supporting method is highly appreciate. A new technique to measure absolute properties should include analysis of the measurement principle and the measurement system. Unfortunately, the presented manuscript does not meet the requirements, neither in content nor in style, and therefore needs fundamental revision before publication.

 

It is proposed to substitute the common procedure according to the standards by a method using heating and cooling cycles, avoiding the blank measurement and the isothermals at the beginning and the end of the measurement. This requires an analysis why the standards recommend the blank. In the used calorimeter model (non explicitly mentioned), the blank can be neglected by using a heating and cooling cycle, since it is the same. In this model, the blank can be also neglected because it should be the same for sapphire and the sample. Due to the presence of temperature gradients in the scanning calorimeter, the blank was not neglected in the classical method, although an additional measurement increase the statistical error. In fact, each DSC furnace has an individual temperature profile depending on the cooling and ambient conditions, temperature, heating rate, gas and crucible. The authors implicitly claim that this temperature profile is symmetrical in heating and cooling. This is highly doubtful and has to be proven. Of course, the differences are minimal near room temperature. However, at other temperatures this is probably completely different. Therefore, a temperature dependence of the asymmetry of the temperature profile between heating and cooling is required. This includes a larger temperature range of the measurements.

 

In this context, I can classify the proposed method as a kind of saw-tooth temperature modulation The amplitude is about 25 K and only the first period are analysed. Wunderlich and/or Andosch studied such kind of modulation from different viewpoints approximately 25 to 30 years ago. The results are published in many papers. The method to determine the heat capacity remains me of a paper obtained by Schenker at al. in 1997. The work of Mathot and Pijpers (e.g. 1989) should be also evaluated.

 

A very positive point of the paper is that liquids are also investigated. This is an extremely  complicated task for heat capacity measurements in disk type DSC because of the influence of the gas volume and potential condensation. The low temperature and small range helps here. However, this requires information about the volume of gas remaining in the crucible. In my opinion, the liquid materials should be characterized with respect to their possible contamination.

 

It must be distinguished between absolute and relative experimental errors (when using a relative measurement technique such as DSC). This is only possible with heat capacity standard materials (as it is available e.g. from NIST) and a statistical investigation with variation the relevant measurement parameters. Such investigations were not supplied.

 

I think my major points are clear enough to help the authors revise the manuscript. Therefore, I will not go into the details. Just one comment on style. Partially, I had the feeling of reading a lecture manuscript for undergraduate students.

 

A final remark reference 1 and 18: The basis of the statements and data made there is not easy to verify. Such references should be avoid in scientific publications.

Author Response

Response to reviewer 3 tries to follow the paragraphs in the review.

1. The calorimeter used is mentioned in the first sentence of the experimental section. The instrument used was a TA Instruments Model 250 DSC. As to different uncertainties for heating and cooling rates, that is baselines, the differences implied by the specifications of the manufacture, mentioned in the text, are negligible, certainly over the modest temperature ranges of the experiments.  This is one of excellent features of the TA Tzero baseline control.
2. The reviewer seems to be asking for an explanation of the instrument’s design. As stated in the paper the basic premise of the reported work is that given the method and repeated measurements, the results agree with literature values in the 1-2 % ranges described with uncertainties in the reported values in the same range. These numerical results in agreement with literature values are the justification for using the method.  A commercial instrument was used in the manner described with results reported.  It does not seem necessary to present the principles of the TA DSC instrument. 
3. A large sample size does reduce problems with vaporization and measuring the c_p of two phases. Preliminary experiments showed that smaller samples in the Tzero sample pans would often give inconsistent results. Moving to larger samples that came close to filling the pans provided much more reproducible results.  Also as the heat capacities of liquids are generally small, certainly compared to solids, the larger sample provided greater detectable signals. The purity of the chemicals used is now in the Materials table.
4. While the material in reference [1] might not be so immediately available as a J Am Chem Soc reference, it is still valid. Reference [18], the Engineer’s Handbook is readily available online with selected and reviewed data with much of it coming from NIST.
5. The authors would have appreciated more clarity on the remark that the manuscript does not meet either the content or style requirements. The manuscript was written to provide clarity for the reader.

Round 2

Reviewer 1 Report

The authors have adequately addressed my concerns with the manuscript.

Author Response

The author's thank the reviewer for thoughtful comments.

Reviewer 3 Report

I accept most of the responses of the authors. However, many of my technical points are not mentioned in their answer.

The first point relates to the novelty of the presented method. In fact, it is a kind of saw-tooth modulation technique that has been extensively discussed by Wunderlich and Androsch and which the Schenker et al. discussed SSADSC is very similar (Thermochimica Acta 304/305 (1997) 219-228 and J. Thermal. Anal. 49 (1997) 1097-1104). The work of these authors must be acknowledged, the distinction from SSADSC must be clearly stated and the title should be changed accordingly.

The second point concerns the temperature range. The authors have shown that the presented technique appears to work in the vicinity of room temperature. More is not demonstrated. This means that the title and the text in the manuscript should also be changed here. The wording "over small temperature ranges" is misleading. Because it could also be a small area around 400 °C, which is not shown.

Author Response

Response to Reviewer 3 second comments.

Reference to using the method for measurements over a temperature range as implied in the title has been removed.  The comment in the text on the possibility of making measurements over short temperature ranges has been expanded noting, however, there is a natural limit dictated by the ability of the sample pan to contain vapor leaks as higher temperatures increase the vapor pressure.

On the issue of incorporating a discussion of modulated DSC using whatever waveform, is moot and not pertinent to our point of simply using an existing instrument in a normal manner.   Yes, modulated DSC in the manner suggested by the reviewer can likely improve the “precision” and perhaps ease of detecting heat capacity, but a review of such instrumentation and methods is not the aim of our contribution.  We therefore feel our presentation of the simple use of an existing instrument using common methods is sufficient for the reader to either accept or dismiss our method as useful.  Other than addressing the concern about perhaps too enthusiastically endorsing the method as one to measure heat capacity temperature dependence, we feel no further change of the manuscript is warranted.

Round 3

Reviewer 3 Report

The remark on modulated DSC and especially on the Stady-State technique refers to the fact that this method is identical in content to the presented technique in the temperature signal and in the evaluation of the heat flow signal. I do not consider the mention of this fact as a matter of form but as a necessity in a methodical scientific work. 

I interpret the authors' answers as meaning that they understand the manuscript as a didactic paper. In this case, it should be published in an appropriate journal . 

Author Response

We have removed the word novel from the title and added a brief comparison of our article to the suggested Thermochimica Acta one.