Please rate your degree of confidence in each of the following study components (0-10 scale; 0 = fundamentally flawed; 1=lowest confidence; 5=medium confidence; 10=highest confidence). An overall confidence score will be calculated as the geometric mean of your three ratings.

Answer Explanations

• 0.00

    View answer

  Expert 4

  	Confidence Rating
  Methods	0 (fundamentally flawed)
  Results	0 (fundamentally flawed)
  Discussion/Conclusions	1 (lowest confidence)

  a)        Incorrect/vague definition of microplastics: Microplastics are particles, fragments or fibers with distinct physical shapes and boundaries/edges. The images provided do not show that the so-called “agglomerations” were microplastics, as there are no clear edges or shapes (as compared to fibers shown in Figure 7). Some of these “agglomerations” likely contained partially dissolved water-soluble film, but they were not microplastics. 
  
  b)        Lack of environmental relevance: Even if microplastics were formed under the experimental conditions, the lack of environmental relevance makes any conclusions not applicable to the real world. If incompletely dissolved film remained under the used experimental conditions, one may expect it  to be degraded by microorganisms after the laundry water is released into the sewer system and treated at a wastewater treatment plant. The time interval used is too short, and the experimental conditions have little representation of the real world. The conclusion that such detergent pods contribute to the formation of microplastics is not substantiated with the experiments, the methods used, and the data presented.  
• 4.93

    View answer

  Expert 3

  	Confidence Rating
  Methods	5 (medium confidence)
  Results	6
  Discussion/Conclusions	4

   Methods: 
  - Are the methods appropriate for the study design and study objectives ? - The methods are appopriate, but other methods like GPC, DLS etc. should have been performed as well. Therefore just medium confidence.
  - Are the study methods (including theory/applicability/modelling) reported in sufficient detail to allow for their replicability or reproducibility? - Yes, it would be easy to repeat the experiments with the information given.

  - Are statistical analyses, controls, sampling mechanism, and statistical reporting (e.g., P-values, CIs, effect sizes) appropriate and well described? - no
  
   

  Results:

  - Is the results presentation, including the number of tables and figures, appropriate to best present the study findings? - Yes, for the experiments conducted, there is enough information.

  - Are additional sub-analyses or statistical measures needed (e.g., reporting of CIs, effect sizes, sensitivity analyses)?  - as outlined above, other experiments should have been conducted as well. repeated experiments and statistical analysis should have been conducted.
  
   

  Discussion/Conclusions:

  - Is the interpretation of results and study conclusions supported by the data and the study design? - Yes, in principle all interpretation points are supported by the data.

  - Have the authors clearly emphasized the limitations of their study/theory/methods/argument?  - No, the study is presented as if there is no limitation.
• 2.62

    View answer

  Expert 2

  	Confidence Rating
  Methods	2
  Results	3
  Discussion/Conclusions	3

  Regarding 1) the experimental work, I found its design to be poor. 
  
  

  1a) First, it is unclear if pH was controlled. Moreover, there were likely residual components in the treated pods that can alter pH (likely increase pH). Water quality parameters (e.g., pH and conductivity) should have been measured before and after dissolution in both tap and deionized water, and for the latter, if the properties changed with time. 
  
  

  1b) The absence of any controls to measure the residual minerals occurring in tap water after evaporation is a major flaw, thereby questioning the validity of the mass balance results for the dissolved pods involving tap water (Table 1 / Fig 2). In addition, it appears that adequate controls are missing for the microscopic image samples (e.g., washing cycles prepared in the absence of pods: water and water + detergent), noting that the washing machine procedure was missing in the Experimental section.
  
  

  1c) It was unclear if experimental replication was used. Moreover, given that the same data was presented in Table 1 and Fig 2 (which is poor presentation practice), the latter but not the former contain error bars. Without error bars (replication) and any statistical analysis, it is unclear if any results measured are statistically significant. Does the experimental procedure support the presentation of results to 3-4 significant figures (per Table 1)?
  
  

  1d) It is unclear whether the drying procedure (353K overnight) removed all of the water during drying, for the work-up of the mass balance measurements. Moreover, it is unclear whether the PVA-rich pod material, being hydrophilic, retained water molecules of hydration. This possible occurrence should have been checked (e.g., via TGA).
  
  

  1e) Did the drying procedure impose any changes in the molecular structure of the pods’ block copolymer: polyvinyl alcohol (PVOH) -polyvinyl acetate (PVAc)?
  
  

  1f) Given the low sensitivity of the mass balance measurements and microscopic analysis, steps should have been taken to remove any microscopic particles that may have occurred in the water sources, e.g., via filtration. 
  
  

  1g) Did the treatment of pod film with acetone produce any artifacts, such as surface modification or dissolution of the PVOH-PVAc film?
  
  

  2. Regarding whether the experimental procedures are relevant to conditions experienced during machine washing of clothes:
  
  

  2a) it is not clear if the conditions used for the mass change tests was reflective of the conditions that would be experienced by detergent pods. For example, does the concentration of pod film per unit volume of water employed in the test relate to the concentration experienced in a washing machine? Does the pH and ionic strength of the water used in the test agree with the water chemistry conditions experienced during clothes washing? Are there catalytic materials in the pods’ liquid contents that accelerate degradation in water that are absent in the tests? Does the temperature employed correspond to warm or hot water washing machine cycles used? Did the time employed for dissolution in the experimental work relate to the residence time experienced by pods in a washing machine?
  
  

  2b) The reader is not given any information on the washing machine tests (per the microscopic results: Figs 6 and 7) to determine whether the experimental protocol was reflective of commonly used conditions.
  
  

  3. The experimental procedure is quite lacking in detail regarding a) properties of tap water; b) procedure for mass change experiments: time for dissolution of film in water; c) no experimental procedures were given for sample preparation for the microscopic analysis (Figs 6 and 7); d) no experimental conditions were given for the FTIR and microscopic instrumentation (in the case of FTIR) or procedures.
  
  4) Regarding interpretation of data: 
  
  

  4a)  No strong proof was given to show that microplastics (MP) formed from the residual films after dissolution in water. It is unclear whether the features depicted in Figs 5 and 6 are directly related to PVOH-PVAc films and no legible length scales were provided in the images. In other words, the main goal of the paper, determining the potential of pods for serving as a source of MP, was not achieved. The absence of experimental detail and likely proper controls (as discussed above) contribute to the uncertainty. It is not clear whether steps were taken to remove particles that would appear in water sources or in the dishwater treatments.
  
  

  4b) Because the contribution of tap water minerals to mass was not taken directly into account, results in Table 1 / Fig 2 involving tap water are not sufficiently robust to allow for conclusions to be drawn.
  
  

  4c) The changes observed in the FTIR hydroxyl and carbonyl indices are counter-intuitive. Noting that hydrolysis of neat PVOH does not occur but does occur for PVAc (via cleavage of the latter’s ester bonds, releasing acetic acid and forming -OH on the polymer backbone), one would expect that the carbonyl index would decrease and the hydroxyl index would increase. A plausible explanation for these results was not provided. Without the explanation, the results produce more questions than answers.
  
  

  4d) It is unclear why FTIR spectral bands employed in the hydroxy and carbonyl indices per Figs 4 and 5 differ from the frequency values listed in Table 2. The inclusion of spectral bands for neat PVOH and neat PVAc, or citing of relevant references, would have strengthened the FTIR analysis. For Table 2, which spectral bands relate to PVOH and which to PVAc?
  
  
• 0.00

    View answer

  Expert 5

  	Confidence Rating
  Methods	1 (lowest confidence)
  Results	1 (lowest confidence)
  Discussion/Conclusions	0 (fundamentally flawed)

  Methods:

  - Are the methods appropriate for the study design and study objectives?

  - Are the study methods (including theory/applicability/modelling) reported in sufficient detail to allow for their replicability or reproducibility?

  
  

  There are a number of experimental details that are missing to enable proper reproducibility:
  
  

  i) Tap water - what is meant by ‘tap water’? Tap water can vary considerably even from tap-to-tap (depending upon local plumbing), and of course from watershed-to-watershed etc. What should have been done would have been a mineral analysis, or a standard solution of a tap water mimic with specific minerals added in specific concentrations. What is the pH of the tap water? 

  
  

  ii) For the dissolution of pouch film in 500 mL of water, was just one pouch film weighed, dissolved and weighed? Not stated. Since the authors then write that they evaporate just 50 mL of solution (10% of the 500 mL of initial solution), were the dry mass amounts (third column, Table 1) multiplied by 10 to compare to the initial mass values (second column, Table 1)? Probably, but I am not sure what happened here, and clarification is needed. 

       We are also missing a number of blank masses: What is the mass of the dried sample of an equivalent volume of tap water with no pod film added? Those dried mineral salts would have a mass. I would also run a blank sample using distilled water, just to check that the dry mass was close to negligible. 

  
  

  iii) The FTIR spectra are hard to see, even when the PDF was blown up on a large monitor, particularly as the distilled water-treated sample is drawn using a dashed line. The differences the authors are analyzing are subtle and nuanced. The feature at ~3300 cm-1 certainly corresponds to R-OH (in polyvinyl alcohol), but can also comprise hydrogen-bonded residual water. As a result, the ratios shown in Figure 4, their hydroxyl indices, are probably unreliable. The evaporation step of air-drying at 313 K will not drive off water efficiently, and subsequent handling of these deliquescent samples could capture water from the air. 

       The ratios of Figure 5, the carbonyl indices, are also difficult to quantitatively measure since the feature at 1450 cm-1, which the author correctly ascribes to C-H bending modes, is noisy and not easy to distinguish. I would take these quantitative values very lightly. 

  
  

  iv) The details for the actual FTIR spectroscopy experiment are also not defined. Is this transmission mode or reflection? If it is transmission, is it taken on a salt plate (if so, which?) or ground with KBr for example? No information is given. 

  
  

  v) The optical images of Figures 6 and 7 lack controls, a readable scale bar (what is the number of the red scale bar?), sufficient experimental information (what clothes were washed, and were they ‘dirty’ in a quantifiable way?) and have been cherry-picked (are the images representative of the entire sample?). We also do not know what the surface of the slide is - glass? Quartz? Is that glass/quartz freshly cleaned? Was the glass/quartz cleaned with, for instance piranha solution (H2SO4:H2O2) before analysis? 

       We also do not know how the microscopy slides with samples were prepared. How much of the aqueous sample was placed on each slide? Was it then dried? And how? 

     

  - Are statistical analyses, controls, sampling mechanism, and statistical reporting (e.g., P-values, CIs, effect sizes) appropriate and well described?

  
  

  No. Standard deviations are indicated in Table 1 (and I assume as well in Figure 2), but number of replicates (and type of replicates - from the same sample, or entirely new samples?) is not indicated. 

  
  

  Standard deviations (I assume!) are also indicated in Figures 4 and 5, but are not defined as such, and the number of replicates is not provided. 

  
  

  
  

  Results:

  - Is the results presentation, including the number of tables and figures, appropriate to best present the study findings?
  - Are additional sub-analyses or statistical measures needed (e.g., reporting of CIs, effect sizes, sensitivity analyses)?

  
  

  These questions are difficult to answer since the experimental design is so flawed. The authors provide 7 figures and 2 tables, but they are not particularly informative. 

  
  

  Discussion/Conclusions:

  - Is the interpretation of results and study conclusions supported by the data and the study design?

  
  

  No. The conclusions of the authors with respect to the experiments carried out are as follows:

  
  

  i) The authors observe a mass change upon dissolution of the pod film and subsequent drying. With distilled water, the mass decreases. With tap water, it increases. With respect to the tap water experiment, a blank using only tap water (no pod film) would have told us how much mineral residue we would expect from 50 mL of water. It would have been interesting to see if the pod films sequestered additional minerals, resulting in an even larger mass increase. 

       The authors did not have any real explanation for the decrease observed with the distilled water. They ascribe the decrease to the “possibility of partial degradation of the capsule film under the influence of water”, but where would the degradation products go? Do the authors assume that they are gaseous? More specifically, the polyvinyl acetate could hydrolyze to release acetic acid, which could evaporate during the drying step, leading to a mass decrease. 

   

  ii) For the FTIR spectra, the authors make conclusions about the ratio of polyvinyl alcohol to polyvinyl acetate. Yes, there are some subtle differences in the FTIR spectra, but it is hard to tell from these images whether the changes are significant. They could result from residual water. If acetate hydrolysis is important, then residual acetic acid could complicate the spectra as well. The ratios of Figures 4 and 5 are an attempt to quantify these difficult-to-read spectra and do not provide much detailed information. 

  
  

  iii) The microscopy images are not helpful in any way. Take a drop of any water that is not Millipure water (with a resistance of >18 ohms), and place it on a freshly cleaned glass microscope slide, and you will observe particulate, adventitious grease, and dust. A fairly concentrated solution of a polymer solution when dropped onto a glass slide and then dried will produce a myriad of patterns, with edge effects, coffee ring patterns, etc. These patterns are kinetically trapped and are not representative of the nature of the solutes in solution. 

       In addition, the claims of identifying micelles using low resolution optical microscopy is nonsensical as micelles are tiny (measured in nanometers) and certainly orders of magnitude smaller than what could be discerned optically. The samples derived from the washing machine would be expected to be especially heterogeneous as there is undefined dirt, fibers, etc. It is impossible to make any conclusions from these images. 

  
  So, to summarize, do the authors demonstrate that "the sludge in pipes can be observed after the usage of such pods", and that "Based on the results, it is clear that detergent pods can be a potential source of microplastics"? No, they do not. The lack of evidence is not evidence of absence, but this work is not meaningful or helpful in answering these questions. 

  
  

  - Have the authors clearly emphasized the limitations of their study/theory/methods/argument?

  
  

  No. The statements provided by the authors are surprisingly definitive. No scientific experiment is ever going to be entirely absolute or conclusive, so one must always provide alternative hypotheses, and address the limitations of their work. The authors do not do that. 

  
  
• 2.71

    View answer

  Expert 1

  	Confidence Rating
  Methods	5 (medium confidence)
  Results	2
  Discussion/Conclusions	2

  I have confidence in the methods themselves, even though not all of the methods used are included in the methods section. The methods are, however, not in all cases linked to the aim of the study and do not allow to draw firm conclusions, but in themselves they are sound and sufficiently described in the paper. The observation that not all methods used are included in the methods section, refers mostly to the procedures used for the washing experiments.
  As can be seen above, the results and the discussion/conclusions sections are very weak, incomplete, not substantiated by scientific findings and in some cases misleading. This especially applies to the final conclusion.