CN105844085A - Detection method of initial decomposition temperature of organic peroxides - Google Patents
Detection method of initial decomposition temperature of organic peroxides Download PDFInfo
- Publication number
- CN105844085A CN105844085A CN201610154615.9A CN201610154615A CN105844085A CN 105844085 A CN105844085 A CN 105844085A CN 201610154615 A CN201610154615 A CN 201610154615A CN 105844085 A CN105844085 A CN 105844085A
- Authority
- CN
- China
- Prior art keywords
- organic peroxide
- model
- detection method
- decomposition temperature
- initial decomposition
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16C—COMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
- G16C20/00—Chemoinformatics, i.e. ICT specially adapted for the handling of physicochemical or structural data of chemical particles, elements, compounds or mixtures
- G16C20/10—Analysis or design of chemical reactions, syntheses or processes
Abstract
The invention discloses a detection method of an initial decomposition temperature of organic peroxides. The detection method comprises following steps: (1) collecting normally-used organic peroxides; (2) optimizing structures of the same kind of the organic peroxides and calculating chemical structure parameters thereof; (3) constructing an initial decomposition temperature To prediction module and evaluation of the module; (4) performing internal verification on the constructed module; (5) defining an application range of the module; and (6) performing quick analysis and prediction on the organic peroxides with unknown initial decomposition temperatures. According to the invention, by detecting the unknown initial decomposition temperatures of the organic peroxides by a QSAR module method, a problem of a lack of thermal hazard parameters of organic peroxides is solved; a disadvantage of a high risk coefficient of a traditional method is overcome; and accuracy of the obtained initial decomposition temperature is improved. The detection method is safe and simple to operate, short in the cycle, uniform in forms and low in dependency, and losses of financial and material resources are greatly reduced.
Description
Technical field
The present invention relates to organic peroxide thermal hazard parameter prediction field, specifically, relate to a kind of organic
The detection method of peroxide initial decomposition temperature.
Background technology
Organic peroxide is the most unstable, is heated and easily decomposes or blast, is particularly easy to generation free radical of decomposing,
It is often used as the initiator of catalyst or radical polymerization.Along with three big synthetic material and functional polymer materials
Developing rapidly of the relevant new techniques such as material, the demand of organic peroxide rapidly increases, and researchers constantly research and develop
Safer, the organic peroxide of novel high-activity and super-active is used in macromolecular material as initiator and closes
Cheng Zhong.
Along with the increase of organic peroxide demand, referred one of the use safety problem of organic peroxide
New height.Due to the thermal instability of organic peroxide, occur quick-fried during producing, transporting and use
The accidents such as fried burning are in the news in succession, to the technology for safely applying of organic peroxide and evaluation system the most gradually by
Bring into schedule.
Researcher main throughput thermal instrument carries out thermal instability and the danger of organic peroxide both at home and abroad
The research of property and exploration.Initial decomposition temperature is claimed heat decomposition temperature again, is that organic peroxide one is important
The physical-chemical parameters, its numerical value is closely coupled with the thermal decomposition explosive nature thereof of organic peroxide.Owing to chemistry is anti-
Should be all the function of temperature, reaction rate may be very slow at low temperatures, it is not intended that do not react generation,
The initial decomposition temperature To finding compound is the more effective way that disaster prevention accident occurs.Cause
This, how obtaining hot risk parameters is an extremely urgent task.Exploitation one both can be cost-effective, again
Can quickly obtain the hot issue that the risk parameters needed badly is the research of current organic peroxide.
QSAR is the structure activity relationship of specified amount, is to use mathematical model to describe certain of molecular structure and molecule
Relation between biological activity.Organic peroxide QSAR method is the structural parameters of self and danger own
Dangerous matter associates, and the most both can reduce the danger in research process, can quickly obtain again our needs
Risk parameters, overcome the weak points such as dose is big, length experimental period, safety coefficient are low.
Summary of the invention
Present invention is generally directed to current organic peroxide initial decomposition temperature parameter not enough, and access approaches is both endangered
The present situation of danger difficulty again, it is provided that a kind of simple and reliable and convenient and safe, it is possible to the initial decomposition temperature needed for acquisition
The detection method of degree parameter.This detection method is according to this Basic Chemical Rule of structures shape performance, according to change
The principle that the character of compound is closely related with molecular structures, that seeks between material microstructure and macroscopic property is interior
In quantitative relationship.Used by the internal relation between the experimental data to molecular structural parameter and studied character
Suitably statistical modeling method is associated, and sets up the quantitative relationship mould between molecular structural parameter and physicochemical property
Type.Once establish reliable Quantitative Structure-Property Relationship correlation model, it is only necessary to the structural information of molecule, so that it may
To predict the initial decomposition temperature of organic peroxide that is new or that not yet synthesize with it.This model is applicable to liquid
State D type organic peroxide, is equally generalized in other close similar organic peroxide compounds.
For solving above-mentioned technical problem, the technical solution used in the present invention is as follows:
The detection method of a kind of organic peroxide initial decomposition temperature, comprises the following steps:
(1) collect conventional organic peroxide, and classify；
(2) optimize the structure of of a sort organic peroxide, and calculate its microstructure parameter；
(3) initial decomposition temperature To forecast model and the evaluation of model are built；
(4) model built is carried out internal verification；
(5) scope of application of this model is defined；
(6) organic peroxide of unknown initial decomposition temperature parameter is quickly analyzed and predicted.
In above-mentioned detection method, step (1) described organic peroxide Type is according to the organic mistake of the United Nations
Oxide criteria for classification belongs to liquid D type organic peroxide, UN number 3105 class organic peroxide.
In above-mentioned detection method, the optimization of step (2) described organic peroxide structure is to use quantization
Learning software Gaussian09, computational methods use temperature Functional Theory DFT；Described microstructure parameter refers to choosing
Taking 20 microstructure parameters of 9 kind of 3105 class organic peroxide, 20 microstructure parameters include:
Highest occupied molecular orbital energy, lowest unoccupied molecular orbital energy, molecular weight, energy, dipole moment, dipole moment square, peroxide
Number of keys, molecule hardness, theoretical active oxygen content, peroxide bridge bond distance, bond angle, dihedral angle, equilibrium oxygen,
Height occupies track can be with lowest unoccupied molecular orbital energy sum, highest occupied molecular orbital can be with the difference of lowest unoccupied molecular orbital energy, peroxide
The electric charge of O atom, the difference of two O atom electric charges, molecular volume, concentration, molecule Mean static polarizabilities on key.
In above-mentioned detection method, step (3) is with organic peroxide thermodynamic parameter initial decomposition temperature To
As planning to build the dependent variable of formwork erection type, use the method removed one by one to carry out Variable Selection, find out and show with dependent variable
Write relevant organic peroxide chemistry structural parameters；Wherein To data control oneself the periodical literature delivered or
The reference value of corresponding concentration in Reaxys data base.Specifically include following steps:
A calculates the theoretical parameter of gained and uses SIMCA-P 11.5 software to carry out partial least square method (PLS) recurrence successively
Analyze, build prediction organic peroxide hot risk parameters initial decomposition temperature To model；
B analysis condition is set to the default value of software, use the mode of truncation select front h composition t1, t2 ...,
Th sets up regression model, and h differentiates Q by Cross gain modulation2H determines；
C is as the Q of certain PLS main constituent2When h is more than 0.0975, it is believed that this main constituent is useful, increase composition
The forecast error reducing model is improved significantly by th；
D differentiates Q when accumulative Cross gain modulation2When cum is more than 0.5, it is believed that the model set up has preferably prediction
Reliability；
E comprehensively uses h, Q2Cum, fitting correlation coefficient (R) and significance level inspection (p) carry out the quality of evaluation model,
Therefrom select the correlation coefficient maximum model the strongest with predictive ability as optimal models.
In above-mentioned detection method, step (4) is that the optimal models filtered out carries out internal cross-verification,
Internal cross-verification method is divided into leaving-one method and stays group method, and this model have employed leaving-one method and carried out internal inspection.Tool
Body comprises the steps:
The data of a kind of peroxide in 9 kinds of organic peroxides are extracted out by a at random；
Then b by the data of other 8 kinds of organic peroxides, carries out the foundation of model according to above-mentioned steps a-e, right
Whether it is consistent with known result than the result of prediction；
C repeats step a and step b, and each data in data by the gross are made above-mentioned inspection successively；
Predicting the outcome of d record inspection every time, if accuracy is more than 70%, then illustrates that this Forecasting Methodology may have
Effect.
In above-mentioned detection method, in step (5), the application of this model be according to the United Nations " about
Dangerous Goods Transport recommendation " in, the liquid D type organic peroxide in organic peroxide criteria for classification,
UN number 3105 class organic peroxide.
In above-mentioned detection method, step (6) including: obtains organic according to the method described in above-mentioned steps
The QSAR predictive equation of peroxide hot risk parameters initial decomposition temperature To, collects and arranges to be predicted same
The value of all structured descriptors of class or close organic peroxide, substitutes into predictive equation and calculates and to be predicted have
The initial decomposition temperature To of machine peroxide.
The present invention is to provide the detection method of a kind of organic peroxide thermal hazard: be based on the structure of matter with
The quantitative relationship of himself character, the prediction of the organic peroxide thermal hazard parameter initial decomposition temperature of structure
Model, by the initial decomposition temperature of QSAR model method prediction organic peroxide.Organic with existing
The preparation method of peroxide thermal hazard parameter is compared, and there is advantages that the present invention passes through
The unknown organic peroxide initial decomposition temperature of QSAR model method prediction, solves organic peroxide and initiates
The present situation that decomposition temperature parameter is deficient, overcome traditional means of experiment complex steps, the high deficiency of danger coefficient it
Place, improves the degree of accuracy of the initial decomposition temperature parameter of acquisition.The method safe ready, cycle be short, form
Unified, dependency is low, greatly reduces the loss of financial resources and material resources.
Accompanying drawing explanation
Fig. 1 is the detection method schematic flow sheet of the present invention.
Fig. 2 is the prediction effect figure of the specific embodiment of the invention.
Fig. 3 is that the present invention specifically predicts the outcome comparison diagram.
Below by embodiment, the invention will be further elaborated, but protection scope of the present invention is not limited to
The scope that embodiment represents.
Detailed description of the invention
The principle of the present invention is the quantitative relationship of architectural feature based on organic peroxide and its thermal hazard matter
Predict the initial decomposition temperature To of unknown organic peroxide.It is comprehensive organic peroxide physics and quantum
Microstructure parameter and the forecast model of its thermal hazard initial decomposition temperature To, and it is applied to prediction the unknown
The method of the initial decomposition temperature (To) of organic peroxide.
As it is shown in figure 1, its to be the present invention a kind of evaluates the initial decomposition temperature (To) of unknown organic peroxide
The schematic flow sheet of detection method, this model construction detailed process is:
Collect conventional organic peroxide, carry out screening and being classified.According to the United Nations " about dangerous goods
Thing transport recommendation " in the criteria for classification of organic peroxide, choose 3105 class organic peroxides.
Use the structure of quantum chemistry software Gaussian09 optimum option organic peroxide, and general according to temperature
The computational methods of letter theory DFT calculate its quantum chemistry structural parameters.Detailed process: choosing 9 kind of 3105 class has
20 microstructure parameters of machine peroxide, 20 microstructure parameters of 9 kinds of organic peroxides specifically wrap
Include: highest occupied molecular orbital energy, lowest unoccupied molecular orbital energy, molecular weight, energy, dipole moment, dipole moment square,
Peroxide bridge number, molecule hardness, theoretical active oxygen content, peroxide bridge bond distance, bond angle, dihedral angle, equilibrium oxygen,
Highest occupied molecular orbital can be with lowest unoccupied molecular orbital energy sum, highest occupied molecular orbital can be with the difference of lowest unoccupied molecular orbital energy, mistake
The electric charge of O atom, the difference of two O atom electric charges, molecular volume, concentration, molecule Mean static polarizabilities on oxygen key.
As shown in table 1:
Table 20 microstructure parameters of 1.9 kinds of organic peroxides
aThe single English alphabet of 9 kind of 3105 class organic peroxide routine represents.
Using organic peroxide thermodynamic parameter initial decomposition temperature To as setting up the dependent variable of model, with by
One method removed carries out Variable Selection, finds out the organic peroxide quantum chemistry knot with dependent variable significant correlation
Structure parameter.Detailed process is: the theoretical parameter calculating gained uses SIMCA-P 11.5 software to carry out partially successively
Method of least square (PLS) regression analysis, builds prediction organic peroxide hot risk parameters initial decomposition temperature To
Model.Analysis condition is set to the default value of software, use the mode of truncation select front h composition t1, t2 ...,
Th sets up regression model, and h differentiates Q by Cross gain modulation2H determines.When certain PLS main constituent
Q2When h is more than 0.0975, it is believed that this main constituent is useful, increase composition th to reducing the prediction of model by mistake
Difference improves significantly.When accumulative Cross gain modulation differentiates Q2When cum is more than 0.5, it is believed that set up
Model have preferable predicting reliability.Comprehensively use h, Q2Cum, fitting correlation coefficient (R) and significance water
Flat inspection (p) etc. carrys out the quality of evaluation model.Therefrom select the model that correlation coefficient is maximum and predictive ability is the strongest
As optimal models.Concrete prediction effect such as Fig. 2.
The model built is carried out the internal validation-cross of leaving-one method.Being embodied as step is: have 9 kinds at random
The data of a kind of peroxide in machine peroxide are extracted out.Then with the number of other 8 kinds of organic peroxides
According to, carry out the foundation of model, whether the result of contrast prediction is consistent with known result.Repeat previous step,
Each data in data by the gross are made above-mentioned inspection successively.Predicting the outcome, if correctly of record inspection every time
Rate more than 70%, then illustrates that this Forecasting Methodology may be effectively.The concrete predictive value of this model is pre-with experiment value
Survey Comparative result such as table 2 and Fig. 3.
2.9 kind of 3105 class organic peroxide experiment value of table and predictive value
Mean: be SADT (experiment value) and the meansigma methods of SADT (predictive value)；
Diff.=SADT (experiment value)-SADT (predictive value)；
The standard deviation of SE:SADT predictive value；
SE/Mean: standard deviation and the multiplying power of meansigma methods.
As can be known from Table 2, standard deviation is less than or equal to 3.6% with the multiplying power of meansigma methods, and present method invention model is described
Prediction effect fine, can promote the use in real work.
Claims (9)
1. the detection method of an organic peroxide initial decomposition temperature, it is characterised in that comprise the following steps:
(1) collect conventional organic peroxide, and classify；
(2) optimize of a sort organic peroxide structure, and calculate its microstructure parameter；
(3) initial decomposition temperature To forecast model and the evaluation of model are built；
(4) model built is carried out internal verification；
(5) scope of application of this model is defined；
(6) organic peroxide of unknown initial decomposition temperature parameter is quickly analyzed and predicted.
Detection method the most according to claim 1, it is characterised in that step (1) described organic peroxide is for belong to liquid D type organic peroxide, UN number 3105 class organic peroxide according to the United Nations's organic peroxide criteria for classification.
Detection method the most according to claim 1, it is characterised in that step (2) described optimization of a sort organic peroxide structure is to use quantum chemistry software Gaussian09, and computational methods use temperature Functional Theory DFT；Described microstructure parameter refers to choose 20 microstructure parameters of 9 kind of 3105 class organic peroxide, 20 microstructure parameters include: highest occupied molecular orbital energy, lowest unoccupied molecular orbital energy, molecular weight, energy, dipole moment, dipole moment square, peroxide bridge number, molecule hardness, theoretical active oxygen content, peroxide bridge bond distance, bond angle, dihedral angle, equilibrium oxygen, highest occupied molecular orbital can be with lowest unoccupied molecular orbital energy sum, highest occupied molecular orbital can be with the difference of lowest unoccupied molecular orbital energy, the electric charge of O atom on peroxide bridge, the difference of two O atom electric charges, molecular volume, concentration, molecule Mean static polarizabilities.
Detection method the most according to claim 1, it is characterized in that, step (3) is using organic peroxide thermodynamic parameter initial decomposition temperature To as the dependent variable planning to build formwork erection type, use the method removed one by one to carry out Variable Selection, find out the organic peroxide chemistry structural parameters with dependent variable significant correlation；Wherein To data are controlled oneself the periodical literature delivered or from the reference value of corresponding concentration in Reaxys data base.
Detection method the most according to claim 4, it is characterised in that comprise the steps:
A calculates the theoretical parameter of gained and uses SIMCA-P 11.5 software to carry out partial least square method regression analysis successively, builds prediction organic peroxide hot risk parameters initial decomposition temperature To model；
B analysis condition is set to the default value of software, uses the mode of truncation to select front h composition t1, t2 ..., th to set up regression model, and h differentiates Q by Cross gain modulation2H determines；
C is as the Q of certain PLS main constituent2When h is more than 0.0975, it is believed that this main constituent is useful, increases composition th and the forecast error reducing model is improved significantly；
D differentiates Q when accumulative Cross gain modulation2Cum is more than 0.5
Time, it is believed that the model set up has preferable predicting reliability；
E comprehensively uses h, Q2Cum, fitting correlation coefficient R and significance level inspection p carry out the quality of evaluation model, therefrom select the correlation coefficient maximum model the strongest with predictive ability as optimal models.
Detection method the most according to claim 1, it is characterised in that step (4) for carrying out internal cross-verification to the model filtered out, and this model have employed leaving-one method and carries out internal inspection.
Detection method the most according to claim 6, it is characterised in that comprise the steps:
The data of a kind of peroxide in 9 kinds of organic peroxides are extracted out by a at random；
Then b by the data of other 8 kinds of organic peroxides, carries out the foundation of model according to the step in claim 5, and whether the result of contrast prediction is consistent with known result；
C repeats step a and step b, and each data in data by the gross are made above-mentioned inspection successively；
Predicting the outcome of d record inspection every time, if accuracy is more than 70%, then illustrates that this Forecasting Methodology may be effectively.
Detection method the most according to claim 1, it is characterized in that, in step (5), the scope of application of model is according in the United Nations's " Recommendations on the Transport of Dangerous Goods ", liquid D type organic peroxide in organic peroxide criteria for classification, UN number 3105 class organic peroxide.
Detection method the most according to claim 1, it is characterized in that, step (6) including: obtains the QSAR predictive equation of organic peroxide hot risk parameters initial decomposition temperature To according to the method described in step in the claims 17, collect and arrange the value of all structured descriptors of to be predicted similar or close organic peroxide, substitute into predictive equation and calculate the initial decomposition temperature To of organic peroxide to be predicted.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610154615.9A CN105844085A (en) | 2016-03-17 | 2016-03-17 | Detection method of initial decomposition temperature of organic peroxides |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610154615.9A CN105844085A (en) | 2016-03-17 | 2016-03-17 | Detection method of initial decomposition temperature of organic peroxides |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105844085A true CN105844085A (en) | 2016-08-10 |
Family
ID=56587298
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610154615.9A Pending CN105844085A (en) | 2016-03-17 | 2016-03-17 | Detection method of initial decomposition temperature of organic peroxides |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105844085A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110146539A (en) * | 2019-05-13 | 2019-08-20 | 南京理工大学 | A method of assessment substance pyrolysis minimal decomposition initial temperature |
| CN113588932A (en) * | 2021-07-29 | 2021-11-02 | 哈尔滨理工大学 | Pyrolysis product monitoring device and method based on P-Q regression model of phenolic resin low-temperature pyrolysis gas |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102445369B (en) * | 2011-09-28 | 2013-03-20 | 中山大学 | Sample pretreatment method for testing thermal parameters of organic peroxide |
-
2016
- 2016-03-17 CN CN201610154615.9A patent/CN105844085A/en active Pending
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102445369B (en) * | 2011-09-28 | 2013-03-20 | 中山大学 | Sample pretreatment method for testing thermal parameters of organic peroxide |
Non-Patent Citations (11)
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110146539A (en) * | 2019-05-13 | 2019-08-20 | 南京理工大学 | A method of assessment substance pyrolysis minimal decomposition initial temperature |
| CN110146539B (en) * | 2019-05-13 | 2021-09-10 | 南京理工大学 | Method for evaluating minimum decomposition starting temperature of material thermal decomposition reaction |
| CN113588932A (en) * | 2021-07-29 | 2021-11-02 | 哈尔滨理工大学 | Pyrolysis product monitoring device and method based on P-Q regression model of phenolic resin low-temperature pyrolysis gas |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Peters | Reaction rate theory and rare events | |
| Jagau et al. | Extending quantum chemistry of bound states to electronic resonances | |
| Young | Data-based mechanistic modelling, generalised sensitivity and dominant mode analysis | |
| CN102980972B (en) | Method for determining hot dangerousness of self-reactive chemical substance | |
| CN101339181B (en) | Organic compound explosive characteristic prediction method based on genetic algorithm | |
| Ricardez-Sandoval | Optimal design and control of dynamic systems under uncertainty: A probabilistic approach | |
| Mueller et al. | Polymerization reactor modeling in industry | |
| Wang et al. | Parameters estimation and model discrimination for solid-liquid reactions in batch processes | |
| CN105844085A (en) | Detection method of initial decomposition temperature of organic peroxides | |
| Fang et al. | Hybrid method integrating machine learning and particle swarm optimization for smart chemical process operations | |
| Kanavalau et al. | Robust thermal stability for batch process intensification with model predictive control | |
| Turányi et al. | Reduction of reaction mechanisms | |
| Gong et al. | Considering different streamflow forecast horizons in the quantitative flood risk analysis for a multi-reservoir system | |
| CN114334025A (en) | Construction and verification method of cement clinker calcination environment variable | |
| Kolhapure et al. | PDF simulations of ethylene decomposition in tubular LDPE reactors | |
| Topaloglu et al. | Machine learning-based estimation of energy dissipation capacity of RC shear walls | |
| Zhang et al. | Modeling and simulation of high-pressure urea synthesis loop | |
| Han et al. | In‐silico vinyl chloride suspension polymerization: Kinetic modelling, thermal runaway prediction, and prevention | |
| Ma et al. | Artificial neural network prediction of metastable zone widths in reactive crystallization of lithium carbonate | |
| CN105653795A (en) | Detection method of self-accelerating decomposition temperature of organic peroxide | |
| CN110457721A (en) | Method is determined based on the chemical industry safety operation domain of global parameter sensitivity analysis | |
| Aguiar | Mathematical modeling of the internal surface area of copolymer particles based on elementary gel structures | |
| Nogueira et al. | Continuous polymerization in tubular reactors with prepolymerization: analysis using two‐dimensional phenomenological model and hybrid model with neural networks | |
| Matsunaga et al. | Finite-time stability analysis for resource limited chemical reactions | |
| CN116631523A (en) | Identification method of semi-intermittent reaction thermal behavior based on neural network |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20160810 |